Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 25 Jan 2001 16:12:43 +0100
From:      Nicolas Souchu <nsouch@alcove.fr>
To:        Andrew Gallatin <gallatin@cs.duke.edu>
Cc:        takawata@shidahara1.planet.sci.kobe-u.ac.jp, freebsd-hackers@freebsd.org
Subject:   Re: AMD System Management driver and Newbus
Message-ID:  <20010125161242.E18032@wiliam.alcove-int>
In-Reply-To: <14924.44189.741374.795171@grasshopper.cs.duke.edu>; from gallatin@cs.duke.edu on Fri, Dec 29, 2000 at 02:26:39PM -0500
References:  <Pine.BSF.4.21.0012271119590.4933-100000@theo.forman.homeip.net> <20001228091913.C26574@wiliam.alcove-int> <14923.39062.862177.106870@grasshopper.cs.duke.edu> <20001229094022.C29262@wiliam.alcove-int> <14924.44189.741374.795171@grasshopper.cs.duke.edu>

next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Dec 29, 2000 at 02:26:39PM -0500, Andrew Gallatin wrote:
> 
> I know what chips they are (From the UP1000 Tech. Ref. Manual):
> 
>   The UP1000 has one main SM bus and two sub-SM busses.  Acces to the
>   sub-SM bus is controlled by a multiplexer (MUX) which sits on the main 
>   SM bus. 
>   
>   The SM bus from the Alpha Slot B Module [processor module, akin to a
>   Slot A Athlon] connects the first sub-SM bus; the DIMM EEPROMs are
>   connected to the second sub-SM bus.
>   
>   SM bus addresses for all UP1000 devices are provided in Table 4-3.
>   
>   Table 4-3: SM Bus Address of Each Device:
>   
>   	Device					Address
>   ----------------------------------	-------------------
>   ADM 9240 (system mgmnt unit)		0101 100X
>   
>   ICS9179-06 (Zero delay clock buffer)	1011 001X
>   
>   PCF8574AT (LED Controller)		1010 001X
>   
>   PCF8574AT (I2C bus MUX Controller)	1010 010X

Usually, PCF8574A addresses most significant bits are 0111
and others are programmable. Considering that the documentation
may be wrong the 0x74 and 0x78 chips are certainly PCF8574AT.

>   
>   PCF8582 (MB revision EEPROM)		1010 001X
>   
>   Revision EEPROM on Alpha Slot B	1010 000X
>   Module
>   LM79 (Thermal Detector on Alpha	0101 111X
>   Slot B Module)
>   Dimm0					1010 000X
>   
>   Dimm1					1010 001X
>   
>   Dimm2					1010 010X
> 
> 
> This knowledge doesn't seem to help me much..  Is the I2C bus MUX
> controller common?  Could that be why the "detect" program shows
> device addresses which are different than those given in the table?

Maybe. One way to know this would be to access the chips by forcing
the address and look in their register if we cannot find a kind of
LM79 signature.

> 
> What software would you suggest for dealing with the mux and accessing
> the lm79?

Controlling the mux seems permitted by one of the PCF8574A. But this chip
is only a serial/parallel converter that will allow you to trig the
mux lines. So we have no more info about the mux and how it is connected
to the PCF outputs...

> 
> BTW, do I need the iic stuff?  All I've got for smb in my config file
> is:

Only if you want to send I2C commands. The alpm driver only support
SMB commands, so it's useless in your case.

> 
> device          smbus           # Bus support, required for smb below.
> device          alpm    1
> device          smb
> 
> 
> Thanks again,

Sorry for the delay.

Nicholas

-- 
Nicolas.Souchu@alcove.fr
Alcôve - Open Source Software Engineer - http://www.alcove.fr


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010125161242.E18032>