Re: Poor disk drive performance with FC3 RC3

[Alan Cox <alan redhat com>]

On Thu, Nov 11, 2004 at 03:09:18PM -0500, David Zeuthen wrote:
> Interesting - I suppose there is no better way to detect media in
> optical drives than polling? IIRC some IDE optical drives emits a signal
> on media change (so we don't have to poll; we just listen to that
> signal) but this is not widely available (since most drives doesn't
> comply with the MMC specs) just like the notification from eject buttons
> is not widely available, correct?

Its not a notification - you have to poll. MMC just added the ability to
pull for pending eject and other MMC features. Its all only in later MMC
and highly optional

> Who controls the power saving policy; the drive firmware or the kernel?

Both.

> Is the jamming of the IDE bus caused by a deficiency in the kernel or is
> that just the true way of the IDE spec?

Its the way IDE works

> How does other operating systems handle this?

Windows seems to have multiple techniques so I guess they have whitelists,
blacklists and strategy routines dependant on MMC features. Eg if they can
poll button state they use that.

Alan