Re: [RFC 0/13] extents and 48bit ext3

On Fri, 9 Jun 2006, Andreas Dilger wrote:
> 
> Umm, and how is this fundamentally different from similar code paths in
> the VFS (e.g. O_DIRECT vs regular writes)?

That's a great argument.

You're arguing that your thing is good by pointing to a known disaster 
area and saying "but that other thing does it too, so it must be good".

O_DIRECT is a piece of crap, and I'm still sorry that I didn't push back 
enough on it. And I _did_ push back on it. But I finally was worn down.

And yes, part of the problem is exactly that it uses _almost_ the same 
paths, but not quite. 

Oh, well.

		Linus