stop using ioends for direct write completions

[Dave Chinner]

On Tue, Feb 02, 2016 at 05:42:37PM +0100, Christoph Hellwig wrote:
> On Tue, Feb 02, 2016 at 10:31:18AM -0500, Brian Foster wrote:
> > FWIW, I don't see any such review comments against the three versions of
> > the "DIO needs an ioend for writes" patch I have in my mailbox, but I
> > easily could have missed something..? But if there wasn't time, then
> > fair enough.
> 
> I'll have to look at the mailboxes, but I remember Dave sending this
> out and complaining.

I don't recall the exact discussion that was had, but at the time it
was an evil that I couldn't see a way of avoiding, and with no other
solution being presented.

ISTR a tie-in with the DAX code, too, but that's gone away now with
the block zeroing during allocation rather than using unwritten
extents and completions for this.

> > If COW is the primary motivator, perhaps we can bundle it with that
> > work?
> 
> The prime motivator is to:
> 
>  (1) avoid a pointless memory allocation
>  (2) avoid a pointless context switch
>  (3) avoid pointless code complexity
> 
> COW is just another case where these show up.

*nod*

Cheers,

Dave.
-- 
Dave Chinner
david at fromorbit.com