1- Your implementation seems to be only for one device..(what about several
disks, some md groups, swap only disks, etc...)
2- To spin up/down ide disks on a specific idle time using hdparm, I think that
you do not need all that rc.spin[up/down] thing, as using
'hdparm -qS timeout /dev/hdx' seems to be enough. (It works like a charm in
my ThinkPad 760XD)
>
> Also, we should not need any code in kernel for spindown, scsi generic
> device should be enough for this one.
>
> Only remaining issue is spinup, which needs to be in kernel; but I
> believe it already is there, anyway (JAZ support).
>
Yes, indeed, please see scsi-idle.c (the userland daemon...).
The rest of the kernel patch is to handle gracefuly the spin
up of _nonremovable_ scsi media (jaz spins down by itself
and spins up on kernel request but it _is_ removable media)
The patch is not mine at all, I've only put it together for 2.1.112,
but I think that this (or something alike) is needed before 2.2...
SCSI gurus out there...
~~~
Manuel J. Galan <manolow@step.es>
Sent on 01-Aug-98.
~~~
I live the way I type; fast, with a lot of mistakes.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html