Re: [PATCH 6/7] mm: Pass pvec directly to find_get_entries

[Matthew Wilcox]

On Mon, Aug 24, 2020 at 06:16:20PM +0200, Jan Kara wrote:
> On Wed 19-08-20 16:05:54, Matthew Wilcox (Oracle) wrote:
> > All callers of find_get_entries() use a pvec, so pass it directly
> > instead of manipulating it in the caller.
> > 
> > Signed-off-by: Matthew Wilcox (Oracle) <willy@xxxxxxxxxxxxx>
> 
> Rather than passing pvec to find_get_entries() and then making everybody
> use it, won't it more consistent WRT the naming to make everybody use
> pagevec_lookup_entries() (which is trivial at this point in the series) and
> then rename find_get_entries() to pagevec_lookup_entries()? I.e., I'd prefer
> if the final function was called pagevec_lookup_entries() because that is
> IMO more consistent with how other functions are named in this area...

It seemed more consistent to me to have everybody using
find_get_entries().  To me the pagevec functions:

1. Are in mm/swap.c (not really sure why)
2. Take pvec as the first argument, not the last
3. Wrap a find_* function

Whereas the find_* functions:

1. Are in mm/filemap.c
2. Take mapping as the first argument
3. Manipulate the XArray directly

We already have functions in filemap which take a pagevec, eg
page_cache_delete_batch() and delete_from_page_cache_batch().

So if we're going to merge the two functions, it seems more natural to
have it in filemap.c and called find_get_entries(), but I'm definitely
open to persuasion on this!