There is currently a lot of work going on to properly redesign
the IO subsystem of Linux, the midlevel SCSI code already has
been pretty much completely rewritten and we've put a lot of effort
into making sure that we are getting rid of the limitations and design
errors of the current generation of SCSI code in Linux.
I am working with Eric to get rid of the remaining "misfeatures"
in the current development code that keep us from supporting
the next generations of parallel, SAA and FC-AL SCSI controllers.
Mike