[lkml]   [1999]   [Jul]   [9]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: Can't sleep less than 20 ms
On Fri, Jul 09, 1999 at 06:46:21PM +0200, Jamie Lokier wrote:
> Rogier Wolff wrote:
> > IA32 should also use HZ=1000. However, the API (now cast in stone)
> > specifies that HZ == 100 when seen from userland. Linus has suggested
> > that the kernel-HZ need not be the same as what is "published" to the
> > user applications.
> HZ == 1000 looks like yet another arbitrary decision. Why is 1000Hz
> good enough yet 100Hz is not?

100 Hz means your minimum scheduling interval is 10 milliseconds,
and *sleep() guarantees *at least* parametrized amount of sleep,
not *exactly* parametrized amount of sleep.

There are also odd reasons in timer code, why HZ of power of two
is best, instead of some "nice" 100 or 1000 ...
(Jiffies to 'struct timeval', I vaguely recall..)

> x86 also takes 10 or more microseconds to service an interrupt (it is
> said because of motherboards not the CPU itself).

How long IRQ processing takes *does* depend upon used chips.
When the IRQ-controller is in modern core-logic chipsets, it
does not need to behave like its ancient ISA-bus-bound precursors.

> Do you really want to spend 1% CPU servicing timer interrupts just so
> _occasionally_ a program can get a short sleep to 2ms accuracy?
> I'd rather see:
> - slow HZ -- low interrupt load
> - accurate timers on demand
> - *precise* timing with accuracy of hardware
> [ And then I'd like to see the accurate timers used for polling network
> cards under load :-) ]

RTC based interrupts ? Highly accurate periodic interrupts,
not re-sceduling based "regular by luck"...

> -- Jamie

/Matti Aarnio <>

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:52    [W:0.109 / U:3.392 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site