linux/arch/x86/include/asm/timer.h
Peter Zijlstra 59eaef78bf x86/tsc: Remodel cyc2ns to use seqcount_latch()
Replace the custom multi-value scheme with the more regular
seqcount_latch() scheme. Along with scrapping a lot of lines, the latch
scheme is better documented and used in more places.

The immediate benefit however is not being limited on the update side.
The current code has a limit where the writers block which is hit by
future changes.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-05-15 10:15:15 +02:00

38 lines
987 B
C

#ifndef _ASM_X86_TIMER_H
#define _ASM_X86_TIMER_H
#include <linux/pm.h>
#include <linux/percpu.h>
#include <linux/interrupt.h>
#include <linux/math64.h>
#define TICK_SIZE (tick_nsec / 1000)
unsigned long long native_sched_clock(void);
extern int recalibrate_cpu_khz(void);
extern int no_timer_check;
extern bool using_native_sched_clock(void);
/*
* We use the full linear equation: f(x) = a + b*x, in order to allow
* a continuous function in the face of dynamic freq changes.
*
* Continuity means that when our frequency changes our slope (b); we want to
* ensure that: f(t) == f'(t), which gives: a + b*t == a' + b'*t.
*
* Without an offset (a) the above would not be possible.
*
* See the comment near cycles_2_ns() for details on how we compute (b).
*/
struct cyc2ns_data {
u32 cyc2ns_mul;
u32 cyc2ns_shift;
u64 cyc2ns_offset;
}; /* 16 bytes */
extern void cyc2ns_read_begin(struct cyc2ns_data *);
extern void cyc2ns_read_end(void);
#endif /* _ASM_X86_TIMER_H */