All of lore.kernel.org
 help / color / mirror / Atom feed
From: Edward Tandi <ed@efix.biz>
To: Timothy Miller <miller@techsource.com>
Cc: joe briggs <jbriggs@briggsmedia.com>,
	Artur Jasowicz <kernel@mousebusiness.com>,
	Brian Jackson <brian@brianandsara.net>,
	Bart SCHELSTRAETE <Bart.SCHELSTRAETE@dhl.com>,
	Kernel mailing list <linux-kernel@vger.kernel.org>
Subject: Re: AMD MP, SMP, Tyan 2466
Date: 26 Jun 2003 00:17:56 +0100	[thread overview]
Message-ID: <1056583075.31265.22.camel@wires.home.biz> (raw)
In-Reply-To: <3EFA2939.2060005@techsource.com>

On Wed, 2003-06-25 at 23:59, Timothy Miller wrote:
> Edward Tandi wrote:
> > 
> > Yes, for SMP mode you absolutely need to use 'registered' RAM. Normal
> > PC2100 ram will work OK with one processor but quickly fails with two (I
> > had the same problems). Apparently, DDR RAM uses one clock edge to
> > transfer in one direction and the opposite edge to transfer back again
> > so the registers do synchronisation between one processor writing to the
> > same location that the other one reads from. That's how it was explained
> > to me anyway.
> > 
> DDR memory works very much like single data rate, except that data is 
> transferred (in whichever direction it's going) on both edges of the 
> clock, thus doubling the transfer rate.  The memory does not switch 
> between reading and writing as you describe it.
> 
> I believe registering is for reliability.  Data is transferred one clock 
> cycle later but reduces signal loading.

Thanks for the clarification. I do not profess to be an expert in the
technology. Two writes or a read+write per clock cycle is close enough
for the purpose of the discussion.

The point I was trying to make is that the registers are there to deal
with an SMP race condition of some sort. Athlon MP motherboards fitted
with two processors will not work properly without 'registered' RAM. I
have hard experience of this and it this experience I am sharing with
someone who is seeing the same symptoms.

Ed-T.



  reply	other threads:[~2003-06-25 23:03 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-06-19 16:37 Crusoe's persistent translation on linux? Samphan Raruenrom
2003-06-19 18:03 ` Vojtech Pavlik
2003-06-19 19:51   ` Samphan Raruenrom
2003-06-20  0:02     ` Nuno Silva
2003-06-20  0:16       ` Linus Torvalds
2003-06-20  2:08         ` Nuno Silva
2003-06-20  9:08         ` Xavier Bestel
2003-06-20  9:33           ` Nick Piggin
2003-06-20 14:08           ` Henning P. Schmiedehausen
2003-06-20 15:38           ` Linus Torvalds
2003-06-20 16:51             ` AMD MP, SMP, Tyan 2466 kernel
2003-06-20 17:08               ` Bart SCHELSTRAETE
2003-06-20 17:09               ` Brian Jackson
2003-06-25 17:37                 ` Artur Jasowicz
2003-06-25 18:51                   ` joe briggs
2003-06-25 18:16                     ` Mike Dresser
2003-07-21 20:52                     ` Artur Jasowicz
2003-07-22  4:02                       ` Jason
2003-07-23 12:19                         ` paterley
2003-07-23 20:03                           ` Joe Briggs
2003-06-25 19:01                   ` joe briggs
2003-06-25 18:56                     ` Edward Tandi
2003-06-25 22:59                       ` Timothy Miller
2003-06-25 23:17                         ` Edward Tandi [this message]
2003-06-25 23:26                           ` Timothy Miller
2003-06-25 23:39                             ` Timothy Miller
2003-06-26  0:29                               ` Edward Tandi
2003-06-25 23:40                             ` Edward Tandi
2003-06-26 15:12                               ` Herbert Poetzl
2003-06-25 23:35                           ` Joel Jaeggli
2003-06-26 12:25                           ` AMD MP, SMP, Tyan 2466, REISERFS I/O error joe briggs
2003-06-26 11:55                             ` Oleg Drokin
2003-06-26 13:37                               ` joe briggs
2003-06-26 23:15                               ` Timothy Miller
2003-06-26 23:48                                 ` Joel Jaeggli
2003-06-27 13:01                                 ` joe briggs
2003-06-20 12:05         ` Crusoe's persistent translation on linux? Samphan Raruenrom
     [not found]     ` <20030619221126.B3287@ucw.cz>
2003-06-23  3:58       ` Crusoe's performance " Samphan Raruenrom
2003-06-23  5:22         ` H. Peter Anvin
2003-06-23  5:40         ` Grzegorz Jaskiewicz
2003-06-23  8:10         ` Vojtech Pavlik
2003-06-23  8:26         ` Vojtech Pavlik
2003-06-23 18:58           ` Samphan Raruenrom
2003-06-24  1:43             ` dean gaudet
2003-06-24  4:33               ` dean gaudet
2003-06-24 22:54                 ` Samphan Raruenrom
2003-06-24 23:06                   ` dean gaudet
2003-06-23 23:46           ` H. Peter Anvin
2003-06-24  0:51         ` H. Peter Anvin
2003-06-26  3:39         ` H. Peter Anvin
2003-06-26  4:22           ` H. Peter Anvin
     [not found]         ` <200306240051.RAA12097@cesium.transmeta.com>
2003-06-26 19:09           ` Samphan Raruenrom

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=1056583075.31265.22.camel@wires.home.biz \
    --to=ed@efix.biz \
    --cc=Bart.SCHELSTRAETE@dhl.com \
    --cc=brian@brianandsara.net \
    --cc=jbriggs@briggsmedia.com \
    --cc=kernel@mousebusiness.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=miller@techsource.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.