All of lore.kernel.org
 help / color / mirror / Atom feed
From: LaMont Jones <lamont@hp.com>
To: Philippe Benard <phi@hpfrcu03.france.hp.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>, LaMont Jones <lamont@hp.com>,
	Stan Sieler <sieler@allegro.com>, Matthew Wilcox <matthew@wil.cx>,
	Jes Sorensen <jes@linuxcare.com>,
	Alan Modra <alan@linuxcare.com.au>,
	John Marvin <jsm@udlkern.fc.hp.com>,
	parisc-linux@thepuffingroup.com, lamont@hp.com
Subject: Re: [parisc-linux] ldcw in __pthread_acquire
Date: Mon, 18 Dec 2000 07:40:33 -0700	[thread overview]
Message-ID: <20001218144036.5BE0B1872C@security.hp.com> (raw)
In-Reply-To: Your message of "Mon, 18 Dec 2000 13:27:43 +0100." <3A3E02BF.DC1D0EB5@hpfrcu03.france.hp.com>

> I think LaMont Jones meant that the spinlock loop (in user space AND in kerne
> space) must implement a load word loop, and issue a load-and-clear word only
> when the lock word looks 'free', this is because the 'write' part of the
> load-and-clear word issue a cache broadcast transaction on the bus, so the
> loop should be on load-word followed by load-clear-word when it looks free,

That was one of the first solutions tried in HP-UX, and it resulted in
processor 4 not getting any time (3 wasn't much better), due to the way
that bus arbitration works (it favors one end of the bus.)

The current semaphore operations in the HP-UX kernel do not use ldcw: they
use stb and ldw in some interesting orders (which break when we get weak
ordering with IA64, but then we'll have a low-cost test-and-set.)

lamont

  reply	other threads:[~2000-12-18 14:37 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2000-12-15 10:12 [parisc-linux] ldcw in __pthread_acquire John Marvin
2000-12-15 11:37 ` Alan Modra
2000-12-15 16:37   ` Matthew Wilcox
2000-12-15 17:32     ` Jes Sorensen
2000-12-16 19:29       ` Matthew Wilcox
2000-12-16 21:58         ` Jes Sorensen
2000-12-17  4:31           ` Alan Modra
2000-12-17  1:22         ` Stan Sieler
2000-12-17  2:38           ` Alan Cox
2000-12-17  4:18             ` LaMont Jones
2000-12-18  0:29             ` Stan Sieler
2000-12-18  0:36               ` Alan Cox
2000-12-18  0:48                 ` Stan Sieler
2000-12-18  0:59                   ` Alan Cox
2000-12-18  4:43                     ` LaMont Jones
2000-12-18 11:53                       ` Alan Cox
2000-12-18 12:27                         ` Philippe Benard
2000-12-18 14:40                           ` LaMont Jones [this message]
2000-12-18 19:44                             ` Stan Sieler
2000-12-18 19:54                               ` Alan Cox
2000-12-18 20:15                                 ` Stan Sieler
2000-12-18 20:44                                 ` LaMont Jones
2000-12-18 21:56                                   ` Alan Cox
2000-12-18 22:26                               ` LaMont Jones
2000-12-18  7:10                 ` Philippe Benard
2000-12-18 12:06                   ` Alan Cox
2000-12-18 14:49                     ` LaMont Jones
2000-12-18 15:59                       ` Matthew Wilcox
  -- strict thread matches above, loose matches on Subject: below --
2000-12-15 10:26 John Marvin

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=20001218144036.5BE0B1872C@security.hp.com \
    --to=lamont@hp.com \
    --cc=alan@linuxcare.com.au \
    --cc=alan@lxorguk.ukuu.org.uk \
    --cc=jes@linuxcare.com \
    --cc=jsm@udlkern.fc.hp.com \
    --cc=matthew@wil.cx \
    --cc=parisc-linux@thepuffingroup.com \
    --cc=phi@hpfrcu03.france.hp.com \
    --cc=sieler@allegro.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.