linux_dsm_epyc7002/arch/arm/kernel/sched_clock.c

185 lines
4.3 KiB
C
Raw Normal View History

ARM: sched_clock: provide common infrastructure for sched_clock() Provide common sched_clock() infrastructure for platforms to use to create a 64-bit ns based sched_clock() implementation from a counter running at a non-variable clock rate. This implementation is based upon maintaining an epoch for the counter and an epoch for the nanosecond time. When we desire a sched_clock() time, we calculate the number of counter ticks since the last epoch update, convert this to nanoseconds and add to the epoch nanoseconds. We regularly refresh these epochs within the counter wrap interval. We perform a similar calculation as above, and store the new epochs. We read and write the epochs in such a way that sched_clock() can easily (and locklessly) detect when an update is in progress, and repeat the loading of these constants when they're known not to be stable. The one caveat is that sched_clock() is not called in the middle of an update. We achieve that by disabling IRQs. Finally, if the clock rate is known at compile time, the counter to ns conversion factors can be specified, allowing sched_clock() to be tightly optimized. We ensure that these factors are correct by providing an initialization function which performs a run-time check. Acked-by: Peter Zijlstra <peterz@infradead.org> Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Tested-by: Will Deacon <will.deacon@arm.com> Tested-by: Mikael Pettersson <mikpe@it.uu.se> Tested-by: Eric Miao <eric.y.miao@gmail.com> Tested-by: Olof Johansson <olof@lixom.net> Tested-by: Jamie Iles <jamie@jamieiles.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-12-16 02:23:07 +07:00
/*
* sched_clock.c: support for extending counters to full 64-bit ns counter
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2 as
* published by the Free Software Foundation.
*/
#include <linux/clocksource.h>
#include <linux/init.h>
#include <linux/jiffies.h>
#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/syscore_ops.h>
ARM: sched_clock: provide common infrastructure for sched_clock() Provide common sched_clock() infrastructure for platforms to use to create a 64-bit ns based sched_clock() implementation from a counter running at a non-variable clock rate. This implementation is based upon maintaining an epoch for the counter and an epoch for the nanosecond time. When we desire a sched_clock() time, we calculate the number of counter ticks since the last epoch update, convert this to nanoseconds and add to the epoch nanoseconds. We regularly refresh these epochs within the counter wrap interval. We perform a similar calculation as above, and store the new epochs. We read and write the epochs in such a way that sched_clock() can easily (and locklessly) detect when an update is in progress, and repeat the loading of these constants when they're known not to be stable. The one caveat is that sched_clock() is not called in the middle of an update. We achieve that by disabling IRQs. Finally, if the clock rate is known at compile time, the counter to ns conversion factors can be specified, allowing sched_clock() to be tightly optimized. We ensure that these factors are correct by providing an initialization function which performs a run-time check. Acked-by: Peter Zijlstra <peterz@infradead.org> Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Tested-by: Will Deacon <will.deacon@arm.com> Tested-by: Mikael Pettersson <mikpe@it.uu.se> Tested-by: Eric Miao <eric.y.miao@gmail.com> Tested-by: Olof Johansson <olof@lixom.net> Tested-by: Jamie Iles <jamie@jamieiles.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-12-16 02:23:07 +07:00
#include <linux/timer.h>
#include <asm/sched_clock.h>
ARM: 7205/2: sched_clock: allow sched_clock to be selected at runtime sched_clock() is yet another blocker on the road to the single image. This patch implements an idea by Russell King: http://www.spinics.net/lists/linux-omap/msg49561.html Instead of asking the platform to implement both sched_clock() itself and the rollover callback, simply register a read() function, and let the ARM code care about sched_clock() itself, the conversion to ns and the rollover. sched_clock() uses this read() function as an indirection to the platform code. If the platform doesn't provide a read(), the code falls back to the jiffy counter (just like the default sched_clock). This allow some simplifications and possibly some footprint gain when multiple platforms are compiled in. Among the drawbacks, the removal of the *_fixed_sched_clock optimization which could negatively impact some platforms (sa1100, tegra, versatile and omap). Tested on 11MPCore, OMAP4 and Tegra. Cc: Imre Kaloz <kaloz@openwrt.org> Cc: Eric Miao <eric.y.miao@gmail.com> Cc: Colin Cross <ccross@android.com> Cc: Erik Gilling <konkers@android.com> Cc: Olof Johansson <olof@lixom.net> Cc: Sascha Hauer <kernel@pengutronix.de> Cc: Alessandro Rubini <rubini@unipv.it> Cc: STEricsson <STEricsson_nomadik_linux@list.st.com> Cc: Lennert Buytenhek <kernel@wantstofly.org> Cc: Ben Dooks <ben-linux@fluff.org> Tested-by: Jamie Iles <jamie@jamieiles.com> Tested-by: Tony Lindgren <tony@atomide.com> Tested-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Linus Walleij <linus.walleij@linaro.org> Acked-by: Nicolas Pitre <nico@linaro.org> Acked-by: Krzysztof Halasa <khc@pm.waw.pl> Acked-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-12-15 18:19:23 +07:00
struct clock_data {
u64 epoch_ns;
u32 epoch_cyc;
u32 epoch_cyc_copy;
u32 mult;
u32 shift;
};
ARM: sched_clock: provide common infrastructure for sched_clock() Provide common sched_clock() infrastructure for platforms to use to create a 64-bit ns based sched_clock() implementation from a counter running at a non-variable clock rate. This implementation is based upon maintaining an epoch for the counter and an epoch for the nanosecond time. When we desire a sched_clock() time, we calculate the number of counter ticks since the last epoch update, convert this to nanoseconds and add to the epoch nanoseconds. We regularly refresh these epochs within the counter wrap interval. We perform a similar calculation as above, and store the new epochs. We read and write the epochs in such a way that sched_clock() can easily (and locklessly) detect when an update is in progress, and repeat the loading of these constants when they're known not to be stable. The one caveat is that sched_clock() is not called in the middle of an update. We achieve that by disabling IRQs. Finally, if the clock rate is known at compile time, the counter to ns conversion factors can be specified, allowing sched_clock() to be tightly optimized. We ensure that these factors are correct by providing an initialization function which performs a run-time check. Acked-by: Peter Zijlstra <peterz@infradead.org> Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Tested-by: Will Deacon <will.deacon@arm.com> Tested-by: Mikael Pettersson <mikpe@it.uu.se> Tested-by: Eric Miao <eric.y.miao@gmail.com> Tested-by: Olof Johansson <olof@lixom.net> Tested-by: Jamie Iles <jamie@jamieiles.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-12-16 02:23:07 +07:00
static void sched_clock_poll(unsigned long wrap_ticks);
static DEFINE_TIMER(sched_clock_timer, sched_clock_poll, 0, 0);
ARM: 7205/2: sched_clock: allow sched_clock to be selected at runtime sched_clock() is yet another blocker on the road to the single image. This patch implements an idea by Russell King: http://www.spinics.net/lists/linux-omap/msg49561.html Instead of asking the platform to implement both sched_clock() itself and the rollover callback, simply register a read() function, and let the ARM code care about sched_clock() itself, the conversion to ns and the rollover. sched_clock() uses this read() function as an indirection to the platform code. If the platform doesn't provide a read(), the code falls back to the jiffy counter (just like the default sched_clock). This allow some simplifications and possibly some footprint gain when multiple platforms are compiled in. Among the drawbacks, the removal of the *_fixed_sched_clock optimization which could negatively impact some platforms (sa1100, tegra, versatile and omap). Tested on 11MPCore, OMAP4 and Tegra. Cc: Imre Kaloz <kaloz@openwrt.org> Cc: Eric Miao <eric.y.miao@gmail.com> Cc: Colin Cross <ccross@android.com> Cc: Erik Gilling <konkers@android.com> Cc: Olof Johansson <olof@lixom.net> Cc: Sascha Hauer <kernel@pengutronix.de> Cc: Alessandro Rubini <rubini@unipv.it> Cc: STEricsson <STEricsson_nomadik_linux@list.st.com> Cc: Lennert Buytenhek <kernel@wantstofly.org> Cc: Ben Dooks <ben-linux@fluff.org> Tested-by: Jamie Iles <jamie@jamieiles.com> Tested-by: Tony Lindgren <tony@atomide.com> Tested-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Linus Walleij <linus.walleij@linaro.org> Acked-by: Nicolas Pitre <nico@linaro.org> Acked-by: Krzysztof Halasa <khc@pm.waw.pl> Acked-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-12-15 18:19:23 +07:00
static struct clock_data cd = {
.mult = NSEC_PER_SEC / HZ,
};
static u32 __read_mostly sched_clock_mask = 0xffffffff;
static u32 notrace jiffy_sched_clock_read(void)
{
return (u32)(jiffies - INITIAL_JIFFIES);
}
static u32 __read_mostly (*read_sched_clock)(void) = jiffy_sched_clock_read;
static inline u64 cyc_to_ns(u64 cyc, u32 mult, u32 shift)
{
return (cyc * mult) >> shift;
}
static unsigned long long cyc_to_sched_clock(u32 cyc, u32 mask)
{
u64 epoch_ns;
u32 epoch_cyc;
/*
* Load the epoch_cyc and epoch_ns atomically. We do this by
* ensuring that we always write epoch_cyc, epoch_ns and
* epoch_cyc_copy in strict order, and read them in strict order.
* If epoch_cyc and epoch_cyc_copy are not equal, then we're in
* the middle of an update, and we should repeat the load.
*/
do {
epoch_cyc = cd.epoch_cyc;
smp_rmb();
epoch_ns = cd.epoch_ns;
smp_rmb();
} while (epoch_cyc != cd.epoch_cyc_copy);
return epoch_ns + cyc_to_ns((cyc - epoch_cyc) & mask, cd.mult, cd.shift);
}
/*
* Atomically update the sched_clock epoch.
*/
static void notrace update_sched_clock(void)
{
unsigned long flags;
u32 cyc;
u64 ns;
cyc = read_sched_clock();
ns = cd.epoch_ns +
cyc_to_ns((cyc - cd.epoch_cyc) & sched_clock_mask,
cd.mult, cd.shift);
/*
* Write epoch_cyc and epoch_ns in a way that the update is
* detectable in cyc_to_fixed_sched_clock().
*/
raw_local_irq_save(flags);
cd.epoch_cyc = cyc;
smp_wmb();
cd.epoch_ns = ns;
smp_wmb();
cd.epoch_cyc_copy = cyc;
raw_local_irq_restore(flags);
}
ARM: sched_clock: provide common infrastructure for sched_clock() Provide common sched_clock() infrastructure for platforms to use to create a 64-bit ns based sched_clock() implementation from a counter running at a non-variable clock rate. This implementation is based upon maintaining an epoch for the counter and an epoch for the nanosecond time. When we desire a sched_clock() time, we calculate the number of counter ticks since the last epoch update, convert this to nanoseconds and add to the epoch nanoseconds. We regularly refresh these epochs within the counter wrap interval. We perform a similar calculation as above, and store the new epochs. We read and write the epochs in such a way that sched_clock() can easily (and locklessly) detect when an update is in progress, and repeat the loading of these constants when they're known not to be stable. The one caveat is that sched_clock() is not called in the middle of an update. We achieve that by disabling IRQs. Finally, if the clock rate is known at compile time, the counter to ns conversion factors can be specified, allowing sched_clock() to be tightly optimized. We ensure that these factors are correct by providing an initialization function which performs a run-time check. Acked-by: Peter Zijlstra <peterz@infradead.org> Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Tested-by: Will Deacon <will.deacon@arm.com> Tested-by: Mikael Pettersson <mikpe@it.uu.se> Tested-by: Eric Miao <eric.y.miao@gmail.com> Tested-by: Olof Johansson <olof@lixom.net> Tested-by: Jamie Iles <jamie@jamieiles.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-12-16 02:23:07 +07:00
static void sched_clock_poll(unsigned long wrap_ticks)
{
mod_timer(&sched_clock_timer, round_jiffies(jiffies + wrap_ticks));
ARM: 7205/2: sched_clock: allow sched_clock to be selected at runtime sched_clock() is yet another blocker on the road to the single image. This patch implements an idea by Russell King: http://www.spinics.net/lists/linux-omap/msg49561.html Instead of asking the platform to implement both sched_clock() itself and the rollover callback, simply register a read() function, and let the ARM code care about sched_clock() itself, the conversion to ns and the rollover. sched_clock() uses this read() function as an indirection to the platform code. If the platform doesn't provide a read(), the code falls back to the jiffy counter (just like the default sched_clock). This allow some simplifications and possibly some footprint gain when multiple platforms are compiled in. Among the drawbacks, the removal of the *_fixed_sched_clock optimization which could negatively impact some platforms (sa1100, tegra, versatile and omap). Tested on 11MPCore, OMAP4 and Tegra. Cc: Imre Kaloz <kaloz@openwrt.org> Cc: Eric Miao <eric.y.miao@gmail.com> Cc: Colin Cross <ccross@android.com> Cc: Erik Gilling <konkers@android.com> Cc: Olof Johansson <olof@lixom.net> Cc: Sascha Hauer <kernel@pengutronix.de> Cc: Alessandro Rubini <rubini@unipv.it> Cc: STEricsson <STEricsson_nomadik_linux@list.st.com> Cc: Lennert Buytenhek <kernel@wantstofly.org> Cc: Ben Dooks <ben-linux@fluff.org> Tested-by: Jamie Iles <jamie@jamieiles.com> Tested-by: Tony Lindgren <tony@atomide.com> Tested-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Linus Walleij <linus.walleij@linaro.org> Acked-by: Nicolas Pitre <nico@linaro.org> Acked-by: Krzysztof Halasa <khc@pm.waw.pl> Acked-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-12-15 18:19:23 +07:00
update_sched_clock();
ARM: sched_clock: provide common infrastructure for sched_clock() Provide common sched_clock() infrastructure for platforms to use to create a 64-bit ns based sched_clock() implementation from a counter running at a non-variable clock rate. This implementation is based upon maintaining an epoch for the counter and an epoch for the nanosecond time. When we desire a sched_clock() time, we calculate the number of counter ticks since the last epoch update, convert this to nanoseconds and add to the epoch nanoseconds. We regularly refresh these epochs within the counter wrap interval. We perform a similar calculation as above, and store the new epochs. We read and write the epochs in such a way that sched_clock() can easily (and locklessly) detect when an update is in progress, and repeat the loading of these constants when they're known not to be stable. The one caveat is that sched_clock() is not called in the middle of an update. We achieve that by disabling IRQs. Finally, if the clock rate is known at compile time, the counter to ns conversion factors can be specified, allowing sched_clock() to be tightly optimized. We ensure that these factors are correct by providing an initialization function which performs a run-time check. Acked-by: Peter Zijlstra <peterz@infradead.org> Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Tested-by: Will Deacon <will.deacon@arm.com> Tested-by: Mikael Pettersson <mikpe@it.uu.se> Tested-by: Eric Miao <eric.y.miao@gmail.com> Tested-by: Olof Johansson <olof@lixom.net> Tested-by: Jamie Iles <jamie@jamieiles.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-12-16 02:23:07 +07:00
}
ARM: 7205/2: sched_clock: allow sched_clock to be selected at runtime sched_clock() is yet another blocker on the road to the single image. This patch implements an idea by Russell King: http://www.spinics.net/lists/linux-omap/msg49561.html Instead of asking the platform to implement both sched_clock() itself and the rollover callback, simply register a read() function, and let the ARM code care about sched_clock() itself, the conversion to ns and the rollover. sched_clock() uses this read() function as an indirection to the platform code. If the platform doesn't provide a read(), the code falls back to the jiffy counter (just like the default sched_clock). This allow some simplifications and possibly some footprint gain when multiple platforms are compiled in. Among the drawbacks, the removal of the *_fixed_sched_clock optimization which could negatively impact some platforms (sa1100, tegra, versatile and omap). Tested on 11MPCore, OMAP4 and Tegra. Cc: Imre Kaloz <kaloz@openwrt.org> Cc: Eric Miao <eric.y.miao@gmail.com> Cc: Colin Cross <ccross@android.com> Cc: Erik Gilling <konkers@android.com> Cc: Olof Johansson <olof@lixom.net> Cc: Sascha Hauer <kernel@pengutronix.de> Cc: Alessandro Rubini <rubini@unipv.it> Cc: STEricsson <STEricsson_nomadik_linux@list.st.com> Cc: Lennert Buytenhek <kernel@wantstofly.org> Cc: Ben Dooks <ben-linux@fluff.org> Tested-by: Jamie Iles <jamie@jamieiles.com> Tested-by: Tony Lindgren <tony@atomide.com> Tested-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Linus Walleij <linus.walleij@linaro.org> Acked-by: Nicolas Pitre <nico@linaro.org> Acked-by: Krzysztof Halasa <khc@pm.waw.pl> Acked-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-12-15 18:19:23 +07:00
void __init setup_sched_clock(u32 (*read)(void), int bits, unsigned long rate)
ARM: sched_clock: provide common infrastructure for sched_clock() Provide common sched_clock() infrastructure for platforms to use to create a 64-bit ns based sched_clock() implementation from a counter running at a non-variable clock rate. This implementation is based upon maintaining an epoch for the counter and an epoch for the nanosecond time. When we desire a sched_clock() time, we calculate the number of counter ticks since the last epoch update, convert this to nanoseconds and add to the epoch nanoseconds. We regularly refresh these epochs within the counter wrap interval. We perform a similar calculation as above, and store the new epochs. We read and write the epochs in such a way that sched_clock() can easily (and locklessly) detect when an update is in progress, and repeat the loading of these constants when they're known not to be stable. The one caveat is that sched_clock() is not called in the middle of an update. We achieve that by disabling IRQs. Finally, if the clock rate is known at compile time, the counter to ns conversion factors can be specified, allowing sched_clock() to be tightly optimized. We ensure that these factors are correct by providing an initialization function which performs a run-time check. Acked-by: Peter Zijlstra <peterz@infradead.org> Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Tested-by: Will Deacon <will.deacon@arm.com> Tested-by: Mikael Pettersson <mikpe@it.uu.se> Tested-by: Eric Miao <eric.y.miao@gmail.com> Tested-by: Olof Johansson <olof@lixom.net> Tested-by: Jamie Iles <jamie@jamieiles.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-12-16 02:23:07 +07:00
{
unsigned long r, w;
u64 res, wrap;
char r_unit;
ARM: 7205/2: sched_clock: allow sched_clock to be selected at runtime sched_clock() is yet another blocker on the road to the single image. This patch implements an idea by Russell King: http://www.spinics.net/lists/linux-omap/msg49561.html Instead of asking the platform to implement both sched_clock() itself and the rollover callback, simply register a read() function, and let the ARM code care about sched_clock() itself, the conversion to ns and the rollover. sched_clock() uses this read() function as an indirection to the platform code. If the platform doesn't provide a read(), the code falls back to the jiffy counter (just like the default sched_clock). This allow some simplifications and possibly some footprint gain when multiple platforms are compiled in. Among the drawbacks, the removal of the *_fixed_sched_clock optimization which could negatively impact some platforms (sa1100, tegra, versatile and omap). Tested on 11MPCore, OMAP4 and Tegra. Cc: Imre Kaloz <kaloz@openwrt.org> Cc: Eric Miao <eric.y.miao@gmail.com> Cc: Colin Cross <ccross@android.com> Cc: Erik Gilling <konkers@android.com> Cc: Olof Johansson <olof@lixom.net> Cc: Sascha Hauer <kernel@pengutronix.de> Cc: Alessandro Rubini <rubini@unipv.it> Cc: STEricsson <STEricsson_nomadik_linux@list.st.com> Cc: Lennert Buytenhek <kernel@wantstofly.org> Cc: Ben Dooks <ben-linux@fluff.org> Tested-by: Jamie Iles <jamie@jamieiles.com> Tested-by: Tony Lindgren <tony@atomide.com> Tested-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Linus Walleij <linus.walleij@linaro.org> Acked-by: Nicolas Pitre <nico@linaro.org> Acked-by: Krzysztof Halasa <khc@pm.waw.pl> Acked-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-12-15 18:19:23 +07:00
BUG_ON(bits > 32);
WARN_ON(!irqs_disabled());
WARN_ON(read_sched_clock != jiffy_sched_clock_read);
read_sched_clock = read;
sched_clock_mask = (1 << bits) - 1;
ARM: sched_clock: provide common infrastructure for sched_clock() Provide common sched_clock() infrastructure for platforms to use to create a 64-bit ns based sched_clock() implementation from a counter running at a non-variable clock rate. This implementation is based upon maintaining an epoch for the counter and an epoch for the nanosecond time. When we desire a sched_clock() time, we calculate the number of counter ticks since the last epoch update, convert this to nanoseconds and add to the epoch nanoseconds. We regularly refresh these epochs within the counter wrap interval. We perform a similar calculation as above, and store the new epochs. We read and write the epochs in such a way that sched_clock() can easily (and locklessly) detect when an update is in progress, and repeat the loading of these constants when they're known not to be stable. The one caveat is that sched_clock() is not called in the middle of an update. We achieve that by disabling IRQs. Finally, if the clock rate is known at compile time, the counter to ns conversion factors can be specified, allowing sched_clock() to be tightly optimized. We ensure that these factors are correct by providing an initialization function which performs a run-time check. Acked-by: Peter Zijlstra <peterz@infradead.org> Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Tested-by: Will Deacon <will.deacon@arm.com> Tested-by: Mikael Pettersson <mikpe@it.uu.se> Tested-by: Eric Miao <eric.y.miao@gmail.com> Tested-by: Olof Johansson <olof@lixom.net> Tested-by: Jamie Iles <jamie@jamieiles.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-12-16 02:23:07 +07:00
/* calculate the mult/shift to convert counter ticks to ns. */
ARM: 7205/2: sched_clock: allow sched_clock to be selected at runtime sched_clock() is yet another blocker on the road to the single image. This patch implements an idea by Russell King: http://www.spinics.net/lists/linux-omap/msg49561.html Instead of asking the platform to implement both sched_clock() itself and the rollover callback, simply register a read() function, and let the ARM code care about sched_clock() itself, the conversion to ns and the rollover. sched_clock() uses this read() function as an indirection to the platform code. If the platform doesn't provide a read(), the code falls back to the jiffy counter (just like the default sched_clock). This allow some simplifications and possibly some footprint gain when multiple platforms are compiled in. Among the drawbacks, the removal of the *_fixed_sched_clock optimization which could negatively impact some platforms (sa1100, tegra, versatile and omap). Tested on 11MPCore, OMAP4 and Tegra. Cc: Imre Kaloz <kaloz@openwrt.org> Cc: Eric Miao <eric.y.miao@gmail.com> Cc: Colin Cross <ccross@android.com> Cc: Erik Gilling <konkers@android.com> Cc: Olof Johansson <olof@lixom.net> Cc: Sascha Hauer <kernel@pengutronix.de> Cc: Alessandro Rubini <rubini@unipv.it> Cc: STEricsson <STEricsson_nomadik_linux@list.st.com> Cc: Lennert Buytenhek <kernel@wantstofly.org> Cc: Ben Dooks <ben-linux@fluff.org> Tested-by: Jamie Iles <jamie@jamieiles.com> Tested-by: Tony Lindgren <tony@atomide.com> Tested-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Linus Walleij <linus.walleij@linaro.org> Acked-by: Nicolas Pitre <nico@linaro.org> Acked-by: Krzysztof Halasa <khc@pm.waw.pl> Acked-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-12-15 18:19:23 +07:00
clocks_calc_mult_shift(&cd.mult, &cd.shift, rate, NSEC_PER_SEC, 0);
ARM: sched_clock: provide common infrastructure for sched_clock() Provide common sched_clock() infrastructure for platforms to use to create a 64-bit ns based sched_clock() implementation from a counter running at a non-variable clock rate. This implementation is based upon maintaining an epoch for the counter and an epoch for the nanosecond time. When we desire a sched_clock() time, we calculate the number of counter ticks since the last epoch update, convert this to nanoseconds and add to the epoch nanoseconds. We regularly refresh these epochs within the counter wrap interval. We perform a similar calculation as above, and store the new epochs. We read and write the epochs in such a way that sched_clock() can easily (and locklessly) detect when an update is in progress, and repeat the loading of these constants when they're known not to be stable. The one caveat is that sched_clock() is not called in the middle of an update. We achieve that by disabling IRQs. Finally, if the clock rate is known at compile time, the counter to ns conversion factors can be specified, allowing sched_clock() to be tightly optimized. We ensure that these factors are correct by providing an initialization function which performs a run-time check. Acked-by: Peter Zijlstra <peterz@infradead.org> Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Tested-by: Will Deacon <will.deacon@arm.com> Tested-by: Mikael Pettersson <mikpe@it.uu.se> Tested-by: Eric Miao <eric.y.miao@gmail.com> Tested-by: Olof Johansson <olof@lixom.net> Tested-by: Jamie Iles <jamie@jamieiles.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-12-16 02:23:07 +07:00
r = rate;
if (r >= 4000000) {
r /= 1000000;
r_unit = 'M';
ARM: 7205/2: sched_clock: allow sched_clock to be selected at runtime sched_clock() is yet another blocker on the road to the single image. This patch implements an idea by Russell King: http://www.spinics.net/lists/linux-omap/msg49561.html Instead of asking the platform to implement both sched_clock() itself and the rollover callback, simply register a read() function, and let the ARM code care about sched_clock() itself, the conversion to ns and the rollover. sched_clock() uses this read() function as an indirection to the platform code. If the platform doesn't provide a read(), the code falls back to the jiffy counter (just like the default sched_clock). This allow some simplifications and possibly some footprint gain when multiple platforms are compiled in. Among the drawbacks, the removal of the *_fixed_sched_clock optimization which could negatively impact some platforms (sa1100, tegra, versatile and omap). Tested on 11MPCore, OMAP4 and Tegra. Cc: Imre Kaloz <kaloz@openwrt.org> Cc: Eric Miao <eric.y.miao@gmail.com> Cc: Colin Cross <ccross@android.com> Cc: Erik Gilling <konkers@android.com> Cc: Olof Johansson <olof@lixom.net> Cc: Sascha Hauer <kernel@pengutronix.de> Cc: Alessandro Rubini <rubini@unipv.it> Cc: STEricsson <STEricsson_nomadik_linux@list.st.com> Cc: Lennert Buytenhek <kernel@wantstofly.org> Cc: Ben Dooks <ben-linux@fluff.org> Tested-by: Jamie Iles <jamie@jamieiles.com> Tested-by: Tony Lindgren <tony@atomide.com> Tested-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Linus Walleij <linus.walleij@linaro.org> Acked-by: Nicolas Pitre <nico@linaro.org> Acked-by: Krzysztof Halasa <khc@pm.waw.pl> Acked-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-12-15 18:19:23 +07:00
} else if (r >= 1000) {
ARM: sched_clock: provide common infrastructure for sched_clock() Provide common sched_clock() infrastructure for platforms to use to create a 64-bit ns based sched_clock() implementation from a counter running at a non-variable clock rate. This implementation is based upon maintaining an epoch for the counter and an epoch for the nanosecond time. When we desire a sched_clock() time, we calculate the number of counter ticks since the last epoch update, convert this to nanoseconds and add to the epoch nanoseconds. We regularly refresh these epochs within the counter wrap interval. We perform a similar calculation as above, and store the new epochs. We read and write the epochs in such a way that sched_clock() can easily (and locklessly) detect when an update is in progress, and repeat the loading of these constants when they're known not to be stable. The one caveat is that sched_clock() is not called in the middle of an update. We achieve that by disabling IRQs. Finally, if the clock rate is known at compile time, the counter to ns conversion factors can be specified, allowing sched_clock() to be tightly optimized. We ensure that these factors are correct by providing an initialization function which performs a run-time check. Acked-by: Peter Zijlstra <peterz@infradead.org> Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Tested-by: Will Deacon <will.deacon@arm.com> Tested-by: Mikael Pettersson <mikpe@it.uu.se> Tested-by: Eric Miao <eric.y.miao@gmail.com> Tested-by: Olof Johansson <olof@lixom.net> Tested-by: Jamie Iles <jamie@jamieiles.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-12-16 02:23:07 +07:00
r /= 1000;
r_unit = 'k';
ARM: 7205/2: sched_clock: allow sched_clock to be selected at runtime sched_clock() is yet another blocker on the road to the single image. This patch implements an idea by Russell King: http://www.spinics.net/lists/linux-omap/msg49561.html Instead of asking the platform to implement both sched_clock() itself and the rollover callback, simply register a read() function, and let the ARM code care about sched_clock() itself, the conversion to ns and the rollover. sched_clock() uses this read() function as an indirection to the platform code. If the platform doesn't provide a read(), the code falls back to the jiffy counter (just like the default sched_clock). This allow some simplifications and possibly some footprint gain when multiple platforms are compiled in. Among the drawbacks, the removal of the *_fixed_sched_clock optimization which could negatively impact some platforms (sa1100, tegra, versatile and omap). Tested on 11MPCore, OMAP4 and Tegra. Cc: Imre Kaloz <kaloz@openwrt.org> Cc: Eric Miao <eric.y.miao@gmail.com> Cc: Colin Cross <ccross@android.com> Cc: Erik Gilling <konkers@android.com> Cc: Olof Johansson <olof@lixom.net> Cc: Sascha Hauer <kernel@pengutronix.de> Cc: Alessandro Rubini <rubini@unipv.it> Cc: STEricsson <STEricsson_nomadik_linux@list.st.com> Cc: Lennert Buytenhek <kernel@wantstofly.org> Cc: Ben Dooks <ben-linux@fluff.org> Tested-by: Jamie Iles <jamie@jamieiles.com> Tested-by: Tony Lindgren <tony@atomide.com> Tested-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Linus Walleij <linus.walleij@linaro.org> Acked-by: Nicolas Pitre <nico@linaro.org> Acked-by: Krzysztof Halasa <khc@pm.waw.pl> Acked-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-12-15 18:19:23 +07:00
} else
r_unit = ' ';
ARM: sched_clock: provide common infrastructure for sched_clock() Provide common sched_clock() infrastructure for platforms to use to create a 64-bit ns based sched_clock() implementation from a counter running at a non-variable clock rate. This implementation is based upon maintaining an epoch for the counter and an epoch for the nanosecond time. When we desire a sched_clock() time, we calculate the number of counter ticks since the last epoch update, convert this to nanoseconds and add to the epoch nanoseconds. We regularly refresh these epochs within the counter wrap interval. We perform a similar calculation as above, and store the new epochs. We read and write the epochs in such a way that sched_clock() can easily (and locklessly) detect when an update is in progress, and repeat the loading of these constants when they're known not to be stable. The one caveat is that sched_clock() is not called in the middle of an update. We achieve that by disabling IRQs. Finally, if the clock rate is known at compile time, the counter to ns conversion factors can be specified, allowing sched_clock() to be tightly optimized. We ensure that these factors are correct by providing an initialization function which performs a run-time check. Acked-by: Peter Zijlstra <peterz@infradead.org> Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Tested-by: Will Deacon <will.deacon@arm.com> Tested-by: Mikael Pettersson <mikpe@it.uu.se> Tested-by: Eric Miao <eric.y.miao@gmail.com> Tested-by: Olof Johansson <olof@lixom.net> Tested-by: Jamie Iles <jamie@jamieiles.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-12-16 02:23:07 +07:00
/* calculate how many ns until we wrap */
ARM: 7205/2: sched_clock: allow sched_clock to be selected at runtime sched_clock() is yet another blocker on the road to the single image. This patch implements an idea by Russell King: http://www.spinics.net/lists/linux-omap/msg49561.html Instead of asking the platform to implement both sched_clock() itself and the rollover callback, simply register a read() function, and let the ARM code care about sched_clock() itself, the conversion to ns and the rollover. sched_clock() uses this read() function as an indirection to the platform code. If the platform doesn't provide a read(), the code falls back to the jiffy counter (just like the default sched_clock). This allow some simplifications and possibly some footprint gain when multiple platforms are compiled in. Among the drawbacks, the removal of the *_fixed_sched_clock optimization which could negatively impact some platforms (sa1100, tegra, versatile and omap). Tested on 11MPCore, OMAP4 and Tegra. Cc: Imre Kaloz <kaloz@openwrt.org> Cc: Eric Miao <eric.y.miao@gmail.com> Cc: Colin Cross <ccross@android.com> Cc: Erik Gilling <konkers@android.com> Cc: Olof Johansson <olof@lixom.net> Cc: Sascha Hauer <kernel@pengutronix.de> Cc: Alessandro Rubini <rubini@unipv.it> Cc: STEricsson <STEricsson_nomadik_linux@list.st.com> Cc: Lennert Buytenhek <kernel@wantstofly.org> Cc: Ben Dooks <ben-linux@fluff.org> Tested-by: Jamie Iles <jamie@jamieiles.com> Tested-by: Tony Lindgren <tony@atomide.com> Tested-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Linus Walleij <linus.walleij@linaro.org> Acked-by: Nicolas Pitre <nico@linaro.org> Acked-by: Krzysztof Halasa <khc@pm.waw.pl> Acked-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-12-15 18:19:23 +07:00
wrap = cyc_to_ns((1ULL << bits) - 1, cd.mult, cd.shift);
ARM: sched_clock: provide common infrastructure for sched_clock() Provide common sched_clock() infrastructure for platforms to use to create a 64-bit ns based sched_clock() implementation from a counter running at a non-variable clock rate. This implementation is based upon maintaining an epoch for the counter and an epoch for the nanosecond time. When we desire a sched_clock() time, we calculate the number of counter ticks since the last epoch update, convert this to nanoseconds and add to the epoch nanoseconds. We regularly refresh these epochs within the counter wrap interval. We perform a similar calculation as above, and store the new epochs. We read and write the epochs in such a way that sched_clock() can easily (and locklessly) detect when an update is in progress, and repeat the loading of these constants when they're known not to be stable. The one caveat is that sched_clock() is not called in the middle of an update. We achieve that by disabling IRQs. Finally, if the clock rate is known at compile time, the counter to ns conversion factors can be specified, allowing sched_clock() to be tightly optimized. We ensure that these factors are correct by providing an initialization function which performs a run-time check. Acked-by: Peter Zijlstra <peterz@infradead.org> Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Tested-by: Will Deacon <will.deacon@arm.com> Tested-by: Mikael Pettersson <mikpe@it.uu.se> Tested-by: Eric Miao <eric.y.miao@gmail.com> Tested-by: Olof Johansson <olof@lixom.net> Tested-by: Jamie Iles <jamie@jamieiles.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-12-16 02:23:07 +07:00
do_div(wrap, NSEC_PER_MSEC);
w = wrap;
/* calculate the ns resolution of this counter */
ARM: 7205/2: sched_clock: allow sched_clock to be selected at runtime sched_clock() is yet another blocker on the road to the single image. This patch implements an idea by Russell King: http://www.spinics.net/lists/linux-omap/msg49561.html Instead of asking the platform to implement both sched_clock() itself and the rollover callback, simply register a read() function, and let the ARM code care about sched_clock() itself, the conversion to ns and the rollover. sched_clock() uses this read() function as an indirection to the platform code. If the platform doesn't provide a read(), the code falls back to the jiffy counter (just like the default sched_clock). This allow some simplifications and possibly some footprint gain when multiple platforms are compiled in. Among the drawbacks, the removal of the *_fixed_sched_clock optimization which could negatively impact some platforms (sa1100, tegra, versatile and omap). Tested on 11MPCore, OMAP4 and Tegra. Cc: Imre Kaloz <kaloz@openwrt.org> Cc: Eric Miao <eric.y.miao@gmail.com> Cc: Colin Cross <ccross@android.com> Cc: Erik Gilling <konkers@android.com> Cc: Olof Johansson <olof@lixom.net> Cc: Sascha Hauer <kernel@pengutronix.de> Cc: Alessandro Rubini <rubini@unipv.it> Cc: STEricsson <STEricsson_nomadik_linux@list.st.com> Cc: Lennert Buytenhek <kernel@wantstofly.org> Cc: Ben Dooks <ben-linux@fluff.org> Tested-by: Jamie Iles <jamie@jamieiles.com> Tested-by: Tony Lindgren <tony@atomide.com> Tested-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Linus Walleij <linus.walleij@linaro.org> Acked-by: Nicolas Pitre <nico@linaro.org> Acked-by: Krzysztof Halasa <khc@pm.waw.pl> Acked-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-12-15 18:19:23 +07:00
res = cyc_to_ns(1ULL, cd.mult, cd.shift);
ARM: sched_clock: provide common infrastructure for sched_clock() Provide common sched_clock() infrastructure for platforms to use to create a 64-bit ns based sched_clock() implementation from a counter running at a non-variable clock rate. This implementation is based upon maintaining an epoch for the counter and an epoch for the nanosecond time. When we desire a sched_clock() time, we calculate the number of counter ticks since the last epoch update, convert this to nanoseconds and add to the epoch nanoseconds. We regularly refresh these epochs within the counter wrap interval. We perform a similar calculation as above, and store the new epochs. We read and write the epochs in such a way that sched_clock() can easily (and locklessly) detect when an update is in progress, and repeat the loading of these constants when they're known not to be stable. The one caveat is that sched_clock() is not called in the middle of an update. We achieve that by disabling IRQs. Finally, if the clock rate is known at compile time, the counter to ns conversion factors can be specified, allowing sched_clock() to be tightly optimized. We ensure that these factors are correct by providing an initialization function which performs a run-time check. Acked-by: Peter Zijlstra <peterz@infradead.org> Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Tested-by: Will Deacon <will.deacon@arm.com> Tested-by: Mikael Pettersson <mikpe@it.uu.se> Tested-by: Eric Miao <eric.y.miao@gmail.com> Tested-by: Olof Johansson <olof@lixom.net> Tested-by: Jamie Iles <jamie@jamieiles.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-12-16 02:23:07 +07:00
pr_info("sched_clock: %u bits at %lu%cHz, resolution %lluns, wraps every %lums\n",
ARM: 7205/2: sched_clock: allow sched_clock to be selected at runtime sched_clock() is yet another blocker on the road to the single image. This patch implements an idea by Russell King: http://www.spinics.net/lists/linux-omap/msg49561.html Instead of asking the platform to implement both sched_clock() itself and the rollover callback, simply register a read() function, and let the ARM code care about sched_clock() itself, the conversion to ns and the rollover. sched_clock() uses this read() function as an indirection to the platform code. If the platform doesn't provide a read(), the code falls back to the jiffy counter (just like the default sched_clock). This allow some simplifications and possibly some footprint gain when multiple platforms are compiled in. Among the drawbacks, the removal of the *_fixed_sched_clock optimization which could negatively impact some platforms (sa1100, tegra, versatile and omap). Tested on 11MPCore, OMAP4 and Tegra. Cc: Imre Kaloz <kaloz@openwrt.org> Cc: Eric Miao <eric.y.miao@gmail.com> Cc: Colin Cross <ccross@android.com> Cc: Erik Gilling <konkers@android.com> Cc: Olof Johansson <olof@lixom.net> Cc: Sascha Hauer <kernel@pengutronix.de> Cc: Alessandro Rubini <rubini@unipv.it> Cc: STEricsson <STEricsson_nomadik_linux@list.st.com> Cc: Lennert Buytenhek <kernel@wantstofly.org> Cc: Ben Dooks <ben-linux@fluff.org> Tested-by: Jamie Iles <jamie@jamieiles.com> Tested-by: Tony Lindgren <tony@atomide.com> Tested-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Linus Walleij <linus.walleij@linaro.org> Acked-by: Nicolas Pitre <nico@linaro.org> Acked-by: Krzysztof Halasa <khc@pm.waw.pl> Acked-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-12-15 18:19:23 +07:00
bits, r, r_unit, res, w);
ARM: sched_clock: provide common infrastructure for sched_clock() Provide common sched_clock() infrastructure for platforms to use to create a 64-bit ns based sched_clock() implementation from a counter running at a non-variable clock rate. This implementation is based upon maintaining an epoch for the counter and an epoch for the nanosecond time. When we desire a sched_clock() time, we calculate the number of counter ticks since the last epoch update, convert this to nanoseconds and add to the epoch nanoseconds. We regularly refresh these epochs within the counter wrap interval. We perform a similar calculation as above, and store the new epochs. We read and write the epochs in such a way that sched_clock() can easily (and locklessly) detect when an update is in progress, and repeat the loading of these constants when they're known not to be stable. The one caveat is that sched_clock() is not called in the middle of an update. We achieve that by disabling IRQs. Finally, if the clock rate is known at compile time, the counter to ns conversion factors can be specified, allowing sched_clock() to be tightly optimized. We ensure that these factors are correct by providing an initialization function which performs a run-time check. Acked-by: Peter Zijlstra <peterz@infradead.org> Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Tested-by: Will Deacon <will.deacon@arm.com> Tested-by: Mikael Pettersson <mikpe@it.uu.se> Tested-by: Eric Miao <eric.y.miao@gmail.com> Tested-by: Olof Johansson <olof@lixom.net> Tested-by: Jamie Iles <jamie@jamieiles.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-12-16 02:23:07 +07:00
/*
* Start the timer to keep sched_clock() properly updated and
* sets the initial epoch.
*/
sched_clock_timer.data = msecs_to_jiffies(w - (w / 10));
ARM: 7205/2: sched_clock: allow sched_clock to be selected at runtime sched_clock() is yet another blocker on the road to the single image. This patch implements an idea by Russell King: http://www.spinics.net/lists/linux-omap/msg49561.html Instead of asking the platform to implement both sched_clock() itself and the rollover callback, simply register a read() function, and let the ARM code care about sched_clock() itself, the conversion to ns and the rollover. sched_clock() uses this read() function as an indirection to the platform code. If the platform doesn't provide a read(), the code falls back to the jiffy counter (just like the default sched_clock). This allow some simplifications and possibly some footprint gain when multiple platforms are compiled in. Among the drawbacks, the removal of the *_fixed_sched_clock optimization which could negatively impact some platforms (sa1100, tegra, versatile and omap). Tested on 11MPCore, OMAP4 and Tegra. Cc: Imre Kaloz <kaloz@openwrt.org> Cc: Eric Miao <eric.y.miao@gmail.com> Cc: Colin Cross <ccross@android.com> Cc: Erik Gilling <konkers@android.com> Cc: Olof Johansson <olof@lixom.net> Cc: Sascha Hauer <kernel@pengutronix.de> Cc: Alessandro Rubini <rubini@unipv.it> Cc: STEricsson <STEricsson_nomadik_linux@list.st.com> Cc: Lennert Buytenhek <kernel@wantstofly.org> Cc: Ben Dooks <ben-linux@fluff.org> Tested-by: Jamie Iles <jamie@jamieiles.com> Tested-by: Tony Lindgren <tony@atomide.com> Tested-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Linus Walleij <linus.walleij@linaro.org> Acked-by: Nicolas Pitre <nico@linaro.org> Acked-by: Krzysztof Halasa <khc@pm.waw.pl> Acked-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-12-15 18:19:23 +07:00
update_sched_clock();
ARM: sched_clock: provide common infrastructure for sched_clock() Provide common sched_clock() infrastructure for platforms to use to create a 64-bit ns based sched_clock() implementation from a counter running at a non-variable clock rate. This implementation is based upon maintaining an epoch for the counter and an epoch for the nanosecond time. When we desire a sched_clock() time, we calculate the number of counter ticks since the last epoch update, convert this to nanoseconds and add to the epoch nanoseconds. We regularly refresh these epochs within the counter wrap interval. We perform a similar calculation as above, and store the new epochs. We read and write the epochs in such a way that sched_clock() can easily (and locklessly) detect when an update is in progress, and repeat the loading of these constants when they're known not to be stable. The one caveat is that sched_clock() is not called in the middle of an update. We achieve that by disabling IRQs. Finally, if the clock rate is known at compile time, the counter to ns conversion factors can be specified, allowing sched_clock() to be tightly optimized. We ensure that these factors are correct by providing an initialization function which performs a run-time check. Acked-by: Peter Zijlstra <peterz@infradead.org> Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Tested-by: Will Deacon <will.deacon@arm.com> Tested-by: Mikael Pettersson <mikpe@it.uu.se> Tested-by: Eric Miao <eric.y.miao@gmail.com> Tested-by: Olof Johansson <olof@lixom.net> Tested-by: Jamie Iles <jamie@jamieiles.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-12-16 02:23:07 +07:00
/*
* Ensure that sched_clock() starts off at 0ns
*/
ARM: 7205/2: sched_clock: allow sched_clock to be selected at runtime sched_clock() is yet another blocker on the road to the single image. This patch implements an idea by Russell King: http://www.spinics.net/lists/linux-omap/msg49561.html Instead of asking the platform to implement both sched_clock() itself and the rollover callback, simply register a read() function, and let the ARM code care about sched_clock() itself, the conversion to ns and the rollover. sched_clock() uses this read() function as an indirection to the platform code. If the platform doesn't provide a read(), the code falls back to the jiffy counter (just like the default sched_clock). This allow some simplifications and possibly some footprint gain when multiple platforms are compiled in. Among the drawbacks, the removal of the *_fixed_sched_clock optimization which could negatively impact some platforms (sa1100, tegra, versatile and omap). Tested on 11MPCore, OMAP4 and Tegra. Cc: Imre Kaloz <kaloz@openwrt.org> Cc: Eric Miao <eric.y.miao@gmail.com> Cc: Colin Cross <ccross@android.com> Cc: Erik Gilling <konkers@android.com> Cc: Olof Johansson <olof@lixom.net> Cc: Sascha Hauer <kernel@pengutronix.de> Cc: Alessandro Rubini <rubini@unipv.it> Cc: STEricsson <STEricsson_nomadik_linux@list.st.com> Cc: Lennert Buytenhek <kernel@wantstofly.org> Cc: Ben Dooks <ben-linux@fluff.org> Tested-by: Jamie Iles <jamie@jamieiles.com> Tested-by: Tony Lindgren <tony@atomide.com> Tested-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Linus Walleij <linus.walleij@linaro.org> Acked-by: Nicolas Pitre <nico@linaro.org> Acked-by: Krzysztof Halasa <khc@pm.waw.pl> Acked-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-12-15 18:19:23 +07:00
cd.epoch_ns = 0;
pr_debug("Registered %pF as sched_clock source\n", read);
}
unsigned long long notrace sched_clock(void)
{
u32 cyc = read_sched_clock();
return cyc_to_sched_clock(cyc, sched_clock_mask);
ARM: sched_clock: provide common infrastructure for sched_clock() Provide common sched_clock() infrastructure for platforms to use to create a 64-bit ns based sched_clock() implementation from a counter running at a non-variable clock rate. This implementation is based upon maintaining an epoch for the counter and an epoch for the nanosecond time. When we desire a sched_clock() time, we calculate the number of counter ticks since the last epoch update, convert this to nanoseconds and add to the epoch nanoseconds. We regularly refresh these epochs within the counter wrap interval. We perform a similar calculation as above, and store the new epochs. We read and write the epochs in such a way that sched_clock() can easily (and locklessly) detect when an update is in progress, and repeat the loading of these constants when they're known not to be stable. The one caveat is that sched_clock() is not called in the middle of an update. We achieve that by disabling IRQs. Finally, if the clock rate is known at compile time, the counter to ns conversion factors can be specified, allowing sched_clock() to be tightly optimized. We ensure that these factors are correct by providing an initialization function which performs a run-time check. Acked-by: Peter Zijlstra <peterz@infradead.org> Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Tested-by: Will Deacon <will.deacon@arm.com> Tested-by: Mikael Pettersson <mikpe@it.uu.se> Tested-by: Eric Miao <eric.y.miao@gmail.com> Tested-by: Olof Johansson <olof@lixom.net> Tested-by: Jamie Iles <jamie@jamieiles.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-12-16 02:23:07 +07:00
}
void __init sched_clock_postinit(void)
{
ARM: 7205/2: sched_clock: allow sched_clock to be selected at runtime sched_clock() is yet another blocker on the road to the single image. This patch implements an idea by Russell King: http://www.spinics.net/lists/linux-omap/msg49561.html Instead of asking the platform to implement both sched_clock() itself and the rollover callback, simply register a read() function, and let the ARM code care about sched_clock() itself, the conversion to ns and the rollover. sched_clock() uses this read() function as an indirection to the platform code. If the platform doesn't provide a read(), the code falls back to the jiffy counter (just like the default sched_clock). This allow some simplifications and possibly some footprint gain when multiple platforms are compiled in. Among the drawbacks, the removal of the *_fixed_sched_clock optimization which could negatively impact some platforms (sa1100, tegra, versatile and omap). Tested on 11MPCore, OMAP4 and Tegra. Cc: Imre Kaloz <kaloz@openwrt.org> Cc: Eric Miao <eric.y.miao@gmail.com> Cc: Colin Cross <ccross@android.com> Cc: Erik Gilling <konkers@android.com> Cc: Olof Johansson <olof@lixom.net> Cc: Sascha Hauer <kernel@pengutronix.de> Cc: Alessandro Rubini <rubini@unipv.it> Cc: STEricsson <STEricsson_nomadik_linux@list.st.com> Cc: Lennert Buytenhek <kernel@wantstofly.org> Cc: Ben Dooks <ben-linux@fluff.org> Tested-by: Jamie Iles <jamie@jamieiles.com> Tested-by: Tony Lindgren <tony@atomide.com> Tested-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Linus Walleij <linus.walleij@linaro.org> Acked-by: Nicolas Pitre <nico@linaro.org> Acked-by: Krzysztof Halasa <khc@pm.waw.pl> Acked-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-12-15 18:19:23 +07:00
/*
* If no sched_clock function has been provided at that point,
* make it the final one one.
*/
if (read_sched_clock == jiffy_sched_clock_read)
setup_sched_clock(jiffy_sched_clock_read, 32, HZ);
sched_clock_poll(sched_clock_timer.data);
}
static int sched_clock_suspend(void)
{
sched_clock_poll(sched_clock_timer.data);
return 0;
}
static struct syscore_ops sched_clock_ops = {
.suspend = sched_clock_suspend,
};
static int __init sched_clock_syscore_init(void)
{
register_syscore_ops(&sched_clock_ops);
return 0;
}
device_initcall(sched_clock_syscore_init);