From: Daniel Phillips <phillips@arcor.de>
To: root@chaos.analogic.com, Linus Torvalds <torvalds@transmeta.com>
Cc: linux-kernel@vger.kernel.org
Subject: Re: HZ, preferably as small as possible
Date: Thu, 18 Jul 2002 15:25:34 +0200 [thread overview]
Message-ID: <E17VBHi-0004oH-00@starship> (raw)
In-Reply-To: <Pine.LNX.3.95.1020718084908.15746A-100000@chaos.analogic.com>
On Thursday 18 July 2002 14:57, Richard B. Johnson wrote:
> On Wed, 17 Jul 2002, Linus Torvalds wrote:
> > On Wed, 17 Jul 2002, Richard B. Johnson wrote:
> > >
> > > It is hardly novel and I can't imagine how Bresenham or whomever
> > > could make such a claim to the obvious. Even the DOS writer(s) used
> > > this technique to get one-second time intervals from the 18.206
> > > ticks/per second.
> >
> > Ehh.. Look at _existing_ linux code to do exactly the same.
> >
> > See update_wall_time_one_tick() and second_overflow() (which does a lot
> > more besides, but it does largely boil down to this "average fractions
> > using basic integer math" thing.
>
> Maybe you see something in the code I don't. In fact, the hardware
> apprears to have been programmed to interrupt at the HZ rate
> using the constant, CLOCK_TICK_RATE, defined in ../asm/timex.h.
> Maybe the hardware can't be programmed to interrupt at HZ so the
> real ticks are adjusted by 'average fractions' code, but it is
> very unclear if this is being done.
>
> Here is a 20 year-old source snippit of some synthetic division
> code used to correct the DOS time by substituting part of INT 08.
Yes, that's the same algorithm all right, and 'synthetic division'
is a much better name for it than the one I used. IMHO, we should be
doing this even when there happens to be an integral relationship
between timer interrupt rate and HZ. It eliminates a bunch of
posturing we'd otherwise be stuck with to explain/work around
restrictions in the choice of intervals. With a little bit of head
scratching it's also possible to add the bookkeeping necessary to
handle varying physical interrupt rates, while still maintaining
the *exact* correct HZ tick count.
Stripping some cruft from your historical example:
SUB WORD PTR [ACCUMULATOR],NUMERATOR
JNC NO_TICK
MOV AX,DIVISOR
ADD WORD PTR [ACCUMULATOR],AX ; Synth div
CALL TICK
NO_TICK:
Pretty hard to beat that for efficiency.
--
Daniel
next prev parent reply other threads:[~2002-07-18 13:21 UTC|newest]
Thread overview: 78+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-07-10 19:59 HZ, preferably as small as possible Grover, Andrew
2002-07-10 21:09 ` george anzinger
2002-07-11 6:03 ` Hannu Savolainen
2002-07-11 7:15 ` george anzinger
2002-07-12 0:36 ` Stevie O
2002-07-12 0:50 ` Thunder from the hill
2002-07-12 0:55 ` Robert Love
2002-07-12 0:58 ` Thunder from the hill
2002-07-12 1:24 ` Alan Cox
2002-07-12 1:37 ` Mark Hahn
2002-07-12 1:09 ` george anzinger
2002-07-12 1:26 ` Roland Dreier
2002-07-12 17:30 ` george anzinger
2002-07-12 1:35 ` Stevie O
2002-07-12 3:01 ` Bernd Eckenfels
2002-07-11 12:54 ` Thunder from the hill
2002-07-11 15:59 ` Martin Dalecki
2002-07-10 21:28 ` Andrew Morton
2002-07-10 21:35 ` Benjamin LaHaise
2002-07-10 21:38 ` Andrew Morton
2002-07-10 21:42 ` Benjamin LaHaise
2002-07-11 2:14 ` CaT
2002-07-11 17:01 ` Martin Dalecki
2002-07-10 22:01 ` Thunder from the hill
2002-07-10 22:09 ` Cort Dougan
2002-07-11 13:36 ` Whoa... (was: Re: HZ, preferably as small as possible) Mark Mielke
2002-07-11 21:08 ` Daniel Phillips
2002-07-10 22:41 ` HZ, preferably as small as possible Thunder from the hill
2002-07-10 22:47 ` Thunder from the hill
2002-07-10 22:49 ` Eli Carter
2002-07-10 23:05 ` Thunder from the hill
2002-07-10 23:08 ` Dave Mielke
2002-07-10 23:13 ` Thunder from the hill
2002-07-10 23:50 ` J.A. Magallon
2002-07-11 0:28 ` Lincoln Dale
2002-07-11 11:35 ` Kasper Dupont
2002-07-11 12:30 ` Alan Cox
2002-07-11 13:37 ` Kasper Dupont
2002-07-11 15:46 ` Alan Cox
2002-07-11 18:51 ` george anzinger
2002-07-15 5:06 ` Linus Torvalds
2002-07-15 16:26 ` Robert Love
2002-07-15 18:56 ` Linus Torvalds
2002-07-15 19:52 ` mbs
2002-07-15 20:01 ` yodaiken
2002-07-16 11:41 ` Vojtech Pavlik
2002-07-17 19:33 ` Daniel Phillips
2002-07-17 20:31 ` Richard B. Johnson
2002-07-17 20:40 ` Daniel Phillips
2002-07-17 21:02 ` Richard B. Johnson
2002-07-17 21:02 ` Linus Torvalds
2002-07-17 21:16 ` Daniel Phillips
2002-07-18 12:57 ` Richard B. Johnson
2002-07-18 13:25 ` Daniel Phillips [this message]
2002-07-18 10:10 ` Kai Henningsen
2002-07-17 20:55 ` Linus Torvalds
-- strict thread matches above, loose matches on Subject: below --
2002-07-11 2:46 Grover, Andrew
2002-07-11 3:01 ` Jeff Garzik
2002-07-11 11:45 ` Alan Cox
2002-07-11 17:08 ` Martin Dalecki
2002-07-11 19:21 ` Albert D. Cahalan
2002-07-16 9:17 ` Kai Henningsen
2002-07-11 20:34 ` Bill Davidsen
2002-07-12 12:01 ` Martin Dalecki
2002-07-15 5:15 ` Linus Torvalds
2002-07-15 6:56 ` Albert D. Cahalan
2002-07-15 8:24 ` Russell King
2002-07-15 15:48 ` David Mosberger
2002-07-15 18:20 ` Albert D. Cahalan
2002-07-15 18:30 ` David Mosberger
2002-07-15 16:07 ` Albert D. Cahalan
2002-07-15 17:06 ` Russell King
2002-07-15 18:43 ` Albert D. Cahalan
2002-07-15 18:53 ` Russell King
2002-07-15 18:50 ` Linus Torvalds
2002-07-15 20:15 ` Albert D. Cahalan
2002-07-15 8:58 ` Dave Mielke
2002-07-11 7:09 ` george anzinger
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=E17VBHi-0004oH-00@starship \
--to=phillips@arcor.de \
--cc=linux-kernel@vger.kernel.org \
--cc=root@chaos.analogic.com \
--cc=torvalds@transmeta.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox