On Tue, Jan 09, 2001 at 09:05:59AM +0100, Dr. Rene Hexel wrote:

 >   From the definitions in the header files (which have been about my
 > only "documentation" in this respect), the driver seems to do it the
 > wrong way here.  However, sifting through the comments in the "other"
 > drivers it seems like there actually may be different wirings on
 > different hardware!
 > 
 >   Well, since the 5000e seems to be consistent with what the header
 > files say, I will swap this for the moment.  We might be needing some
 > more flexible solution if other people report this to be the wrong way
 > 'round for them.

The right thing to do here would be to detect the "swapped channel"
hardware based on subsystem ID.

-- 
        -- Jason R. Thorpe <thorpej@zembu.com>