Re: [PATCH] direct-IO API change

[Andrew Morton <akpm@digeo.com>]

Linus Torvalds wrote:
> 
> On Fri, 4 Oct 2002, Andrew Morton wrote:
> >
> > Because the file handle which we have is for /dev/raw/raw0,
> > not for /dev/hda1.
> >
> > The raw driver binds to major/minor, not a file*.  I considered
> > changing that (change userspace to pass the open fd).  But didn't.
> 
> Ok. I'd really rather have a cleaner internal API and break the raw driver
> for a while, than have a silly API just because the raw driver uses it.

OK - bust it.

> Especially since I thought that O_DIRECT on the regular file (or block
> device) performed about as well as raw does anyway these days? Or is that
> just one of my LSD-induced flashbacks?
> 

Now we're not holding i_sem for O_DIRECT writes to blockdevs,
I don't think the raw driver offers any advantages at all.  It's
a compatibility thing to save people from having to add "|O_DIRECT" to
their source and then typing `ln -s /dev/hda1 /dev/raw/raw0'.

I think we can probably delete the raw driver.  But I've Cc'ed Janet
and Badari to find out why that's wrong.