Miles Lane wrote:
> > > What you want is to run as fast as usual, with all the nice
> > > elevator and write back rules, but with writes to the disk
> > > sustained until there aren't any further dirty blocks.
> > >
> > > This strikes me as perfectly reasonable behaviour for all
> > > removable media and I can see no reason why anyone argues
> > > against it. It's not like there's a performance cost to
> > > simply writing your dirty data when the medium's request
> > > queue empties.
>
> Is there agreement that this is reasonable?
> Who would be the best person to implement this?
> Jamie, since you seem to have some notion of the
> mechanisms that might be used, would you be
> willing to make the case for this new functionality
> to whomever the possible implementer might be?
Well, as long as the implementor is one of the existing fs gurus, I
expect they can clearly see what mechanisms to use. There's writing
dirty blocks when the disk isn't doing anything else. And there's the
other matter of auto-unmounting, disk change sensing etc. The latter is
complicated, but you can get a long way towards fixing data integrity
without it.
But it's not a very interesting a task. Who cares about optimising
floppy writes? Not me, I hardly ever use floppies and when I do I use
mtools anyway.
Last time a discussion on this topic started out simple and rational, it
quickly degenerated into "how can we emulate the Amiga" and "but I don't
want to run X" and "use Gnome/KDE" etc.
I think the few people interested enough to fix the basic stuff got
bored and moved to another thread.
-- Jamie
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
