On Thu, 20 Jun 2002, Wojciech Puchar wrote:

> > > On Thu, Jun 20, 2002 at 03:01:46PM +0200, Wojciech Puchar wrote:
> >
> > I think that an i/o scheduler is the long-term solution to this problem.
> > It won't make 1.6 (it's WAY too late for that), but it's what we should
> > do.
> >
> WAIT?!
>
> do you mean final 1.6 will be left as is now? with these stalls, swapouts
> etc?

What did I say?

I said that an i/o scheduler will not make 1.6. While I am not in releng
(the release engineering group), it seems like a clear call. We have
already cut 1.6, and have not started working on an i/o scheduler. 1.6 has
been much delayed. To further delay it while awaiting code which
realistically will need months (like maybe 6 or more) to stablize (write,
test, see what needs tunning, modify, test, tune, etc.) would be
foolhearty.

I did not say that we would do nothing to make life better for 1.6.
Certainly the patch you mentioned that Chuck thought he'd already
committed should make 1.6 (note: I'm not in releng, I'm just guessing).
Beyond that, well, it's late in the game.

Take care,

Bill