Re: Pagecache zeroing: zero_user_segment, zero_user_segments and zero_user

[Andrew Morton <akpm@linux-foundation.org>]

On Wed, 16 May 2007 14:52:05 +1000 Nick Piggin <nickpiggin@yahoo.com.au> wrote:

> Christoph Lameter wrote:
> > On Tue, 15 May 2007, Andrew Morton wrote:
> > 
> > 
> >>>   The functions above default to KM_USER0 which is also always used when
> >>>   zero_user_page was called except in one single case. We open code that
> >>>   single case to draw attention to the spot.
> >>>
> >>
> >>Dunno.  fwiw, we decided to _not_ embed KM_USER0 in the callee: we have had
> >>some pretty ghastly bugs in the past due to misuse of kmap slots so the
> >>idea was to shove the decision into the caller's face, make them think
> >>about what they're doing
> > 
> > 
> > On the other hand non highmem platforms are burdened with always repeating
> > the same KM_USER0 in every function call. Isnt it enough to know that 
> > standard functions use KM_USER0 for their operations?
> 
> Couldn't that be filtered out inline?

It is - there is no runtime overhead for non-highmem machines.

The problem nowadays is all the developers who don't need, have, compile
for or test on highmem machines.