Am Donnerstag, 28. April 2005 23:14 schrieb David Brownell:
> On Thursday 28 April 2005 2:01 pm, Oliver Neukum wrote:
> > Am Donnerstag, 28. April 2005 17:40 schrieb David Brownell:
> > > On Thursday 28 April 2005 12:13 am, Oliver Neukum wrote:
> > > >
> > > > Please consider scsi. It has no idea about what is going on.
> > >
> > > No more than for example an IDE disk does. Yet somehow
> > > "hdparm -S" lets the drives themselves spin down when
> > > they're idle for a while ... a kind of autosuspend.
> >
> > An IDE disk understands IDE. A scsi driver does not. Commands are
> > just packets of bytes. And we are talking about a much more limited
> > class of devices.
>
> I really don't see what you're saying. If you're arguing
> that nothing in the SCSI stack can just notice -- as the
> IDE disk does in that example -- that no requests have come
> by in ages, then you'd have an argument on your hands.
> (If I wanted to argue something so self-evidently wrong.)
It is not enough to know whether commands have come down the stack.
Scsi devices are more complex than IDE drivers. Many of them have state.
You'll need the real open/close information and that is lost way up in the
stack.
Regards
Oliver
-------------------------------------------------------
SF.Net email is sponsored by: Tell us your software development plans!
Take this survey and enter to win a one-year sub to SourceForge.net
Plus IDC's 2005 look-ahead and a copy of this survey
Click here to start! http://www.idcswdc.com/cgi-bin/survey?id=105hix
_______________________________________________
linux-usb-devel@lists.sourceforge.net
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
