In message: <20040523092611.GF18223@drowsy.duskware.de>
            Martin Husemann <martin@duskware.de> writes:
: On Sun, May 23, 2004 at 11:22:32AM +0200, Martin Husemann wrote:
: > then. (In the wd driver). This would cause any intermediate
: > enable()/disable() calls to not change the power state. But the first enable
: > would be after the first power off after reading the CIS.
: 
: Btw: I had assumed this kind of drives would be smart enough to not make the
: disk spin up when the card is powered up unconditionally, but only after some
: commands. So maybe it's not worth the effort to avoid the CIS related power
: cycle?

IDE devices are never smart. :-) Well, at least in this generation of
technology.  This is also true for the PCMCIA to 2.5" hard disk
adapters.  I never understood why power is turned off after CIS
parsing, and then turned back on for the attchment phase.  It seems
that there's no power savings to be gained by doing this and always
struck me as a bug (but I never had enough patience to fix this bug).

Warner