Re: [PATCH] O_DIRECT for ext3 (2.4.21)

[Andrew Morton]

"Stephen C. Tweedie" <[e-mail]> wrote:
>
> Hi,
> 
> On Sun, 2003-06-15 at 12:01, Roy Sigurd Karlsbakk wrote:
> > hi all
> > 
> > I've been waiting for the official O_DIRECT on ext3 for some time now, so I
> > thought perhaps it's time to get it into 2.4.22 or so. The patch I've used, is
> > the one below (for 2.4.21):
> 
> This is Andrea's patch,

Actually I'm the culprit.

> (like allowing direct IO in journaled data mode --- bad move ---

hmm, OK, it doesn't even vaguely work in journalled mode either...

I think the check should be implemented in (the new) ext3_open().  Because
checking the return from open() is the way in which a good application would
determine whether the underlying fs supports O_DIRECT.

Unfortunately O_DIRECT can also be set with fcntl(F_SETFL), and we seem to
have forgotten to provide a way for the fs to be told about fcntl.