[ltp] Re: Another one bites the dust

[linux-thinkpad@www.bm-soft.com]

This seems like sound advice to me.

Starting with 2.3.x and (now) into 2.4.x, we are trying to unify the
I2C code into a single set of functions.  You might be surprised to
see just how much I2C code duplication there is in the kernel drivers! 
It's especially nasty since most are derived from Simon Vogl's
original I2C project, and so often use his original symbol names which
conflict and cause nasty crashes and hangs.  The point being, that
disabling I2C support is a short-term solution until something gets
unified which Thinkpads depend on (if anything I2C related), then
you'll need to turn it on again... 

Thinkpads are the only animal I know of, though, that are sensitive
enough to being rendered completely inoperable after an install.  I'm
still not certain what the event is which causes the trouble.  I'd
guess that turing I2C off is the safest thing to do, then not using
any of the Lm_sensor modules, and lastly (the riskiest) is to not run
sensors-detect (which touches every device on the I2C bus[ses]).  We
try to make the package as safe as we can.  In fact, there is a module
to access EEPROM's, but it purposely is read-only eventhough r/w
functionaliy is fairly trivial to add. 


Phil

On Tue, Mar 20, 2001 at 09:28:56PM +0100, Friedemann Baitinger wrote:
> 
> Just make sure that you configure your kernel such that I2C is not
> supported at kernel level at all. Make the following all comments by
> prefixing them with a '#" in the first column:
> 
> #
> # I2C support
> #
> CONFIG_I2C=m
> CONFIG_I2C_ALGOBIT=m
> CONFIG_I2C_PHILIPSPAR=m
> CONFIG_I2C_ELV=m
> CONFIG_I2C_VELLEMAN=m
> CONFIG_I2C_ALGOPCF=m
> CONFIG_I2C_ELEKTOR=m
> CONFIG_I2C_CHARDEV=m

-- 
Philip Edelbrock -- IS Manager -- Edge Design, Corvallis, OR
   phil@netroedge.com -- http://www.netroedge.com/~phil
 PGP F16: 01 D2 FD 01 B5 46 F4 F0  3A 8B 9D 7E 14 7F FB 7A

----- The Linux ThinkPad mailing list -----
The linux-thinkpad mailing list home page is at:
http://www.bm-soft.com/~bm/tp_mailing.html