Re: [PATCH] mm/slub: replace alloc_pages with folio_alloc

[Matthew Wilcox]

On Sun, May 29, 2022 at 04:31:07AM +0100, Matthew Wilcox wrote:
> On Sun, May 29, 2022 at 10:58:18AM +0800, Muchun Song wrote:
> > On Sat, May 28, 2022 at 05:27:11PM +0100, Matthew Wilcox wrote:
> > > On Sun, May 29, 2022 at 12:11:58AM +0800, bh1scw@xxxxxxxxx wrote:
> > > > From: Fanjun Kong <bh1scw@xxxxxxxxx>
> > > > 
> > > > This patch will use folio allocation functions for allocating pages.
> > > 
> > > That's not actually a good idea.  folio_alloc() will do the
> > > prep_transhuge_page() step which isn't needed for slab.
> > 
> > You mean folio_alloc() is dedicated for THP allocation?  It is a little
> > surprise to me.  I thought folio_alloc() is just a variant of alloc_page(),
> > which returns a folio struct instead of a page.  Seems like I was wrong.
> > May I ask what made us decide to do this?
> 
> Yeah, the naming isn't great here.  The problem didn't really occur
> to me until I saw this patch, and I don't have a good solution yet.

OK, I have an idea.

None of the sl*b allocators use the page refcount.  So the
atomic operations on it are just a waste of time.  If we add an
alloc_unref_page() to match our free_unref_page(), that'll be enough
difference to stop pepole sending "helpful" patches.  Also, it'll be a
(small?) performance improvement for slab.