kernel-ark/include/asm-ia64/timex.h
hawkes@sgi.com defbb2c929 [IA64] ia64: simplify and fix udelay()
The original ia64 udelay() was simple, but flawed for platforms without
synchronized ITCs:  a preemption and migration to another CPU during the
while-loop likely resulted in too-early termination or very, very
lengthy looping.

The first fix (now in 2.6.15) broke the delay loop into smaller,
non-preemptible chunks, reenabling preemption between the chunks.  This
fix is flawed in that the total udelay is computed to be the sum of just
the non-premptible while-loop pieces, i.e., not counting the time spent
in the interim preemptible periods.  If an interrupt or a migration
occurs during one of these interim periods, then that time is invisible
and only serves to lengthen the effective udelay().

This new fix backs out the current flawed fix and returns to a simple
udelay(), fully preemptible and interruptible.  It implements two simple
alternative udelay() routines:  one a default generic version that uses
ia64_get_itc(), and the other an sn-specific version that uses that
platform's RTC.

Signed-off-by: John Hawkes <hawkes@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2006-02-15 13:37:04 -08:00

43 lines
1.3 KiB
C

#ifndef _ASM_IA64_TIMEX_H
#define _ASM_IA64_TIMEX_H
/*
* Copyright (C) 1998-2001, 2003 Hewlett-Packard Co
* David Mosberger-Tang <davidm@hpl.hp.com>
*/
/*
* 2001/01/18 davidm Removed CLOCK_TICK_RATE. It makes no sense on IA-64.
* Also removed cacheflush_time as it's entirely unused.
*/
#include <asm/intrinsics.h>
#include <asm/processor.h>
typedef unsigned long cycles_t;
extern void (*ia64_udelay)(unsigned long usecs);
/*
* For performance reasons, we don't want to define CLOCK_TICK_TRATE as
* local_cpu_data->itc_rate. Fortunately, we don't have to, either: according to George
* Anzinger, 1/CLOCK_TICK_RATE is taken as the resolution of the timer clock. The time
* calculation assumes that you will use enough of these so that your tick size <= 1/HZ.
* If the calculation shows that your CLOCK_TICK_RATE can not supply exactly 1/HZ ticks,
* the actual value is calculated and used to update the wall clock each jiffie. Setting
* the CLOCK_TICK_RATE to x*HZ insures that the calculation will find no errors. Hence we
* pick a multiple of HZ which gives us a (totally virtual) CLOCK_TICK_RATE of about
* 100MHz.
*/
#define CLOCK_TICK_RATE (HZ * 100000UL)
static inline cycles_t
get_cycles (void)
{
cycles_t ret;
ret = ia64_getreg(_IA64_REG_AR_ITC);
return ret;
}
#endif /* _ASM_IA64_TIMEX_H */