Re: making x86 clear_user not suck, was Re: [PATCH] tmpfs: fix regressions from wider use of ZERO_PAGE

[Matthew Wilcox]

On Tue, Apr 12, 2022 at 06:57:57AM +0200, Christoph Hellwig wrote:
> On Fri, Apr 08, 2022 at 11:08:29PM -0700, Hugh Dickins wrote:
> > > 
> > > Either way I'd rather do this optimization in iov_iter_zero rather
> > > than hiding it in tmpfs.
> > 
> > Let's see what others say.  I think we would all prefer clear_user() to be
> > enhanced, and hack around it neither here in tmpfs nor in iov_iter_zero().
> > But that careful work won't get done by magic, nor by me.
> 
> I agree with that.
> 
> > And iov_iter_zero() has to deal with a wider range of possibilities,
> > when pulling in cache lines of ZERO_PAGE(0) will be less advantageous,
> > than in tmpfs doing a large dd - the case I'm aiming not to regress here
> > (tmpfs has been copying ZERO_PAGE(0) like this for years).
> 
> Maybe.  OTOH I'd hate to have iov_iter_zero not used much because it
> sucks too much.
> 
> So how can we entice someone with the right knowledge to implement a
> decent clear_user for x86?

Apparently that already happened, but it needs finishing up:
https://lore.kernel.org/lkml/Yk9yBcj78mpXOOLL@xxxxxxxxx/