Re: [PATCHv3, RESEND 6/8] khugepaged: Allow to collapse PTE-mapped compound pages

From: "Kirill A. Shutemov" <kirill@shutemov.name>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>,
	Andrea Arcangeli <aarcange@redhat.com>, Zi Yan <ziy@nvidia.com>,
	Yang Shi <yang.shi@linux.alibaba.com>,
	Ralph Campbell <rcampbell@nvidia.com>,
	John Hubbard <jhubbard@nvidia.com>,
	William Kucharski <william.kucharski@oracle.com>,
	linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCHv3, RESEND 6/8] khugepaged: Allow to collapse PTE-mapped compound pages
Date: Thu, 16 Apr 2020 01:59:11 +0300	[thread overview]
Message-ID: <20200415225911.ryybsmdrtwxngi5h@box> (raw)
In-Reply-To: <20200415155259.8be9887c64b1420cfc9b9655@linux-foundation.org>

On Wed, Apr 15, 2020 at 03:52:59PM -0700, Andrew Morton wrote:
> On Thu, 16 Apr 2020 00:52:05 +0300 "Kirill A. Shutemov" <kirill@shutemov.name> wrote:
> 
> > On Wed, Apr 15, 2020 at 02:44:26PM -0700, Andrew Morton wrote:
> > > On Mon, 13 Apr 2020 15:52:18 +0300 "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com> wrote:
> > > 
> > > > We can collapse PTE-mapped compound pages. We only need to avoid
> > > > handling them more than once: lock/unlock page only once if it's present
> > > > in the PMD range multiple times as it handled on compound level. The
> > > > same goes for LRU isolation and putback.
> > > > 
> > > > ...
> > > >
> > > > --- a/mm/khugepaged.c
> > > > +++ b/mm/khugepaged.c
> > > > @@ -515,17 +515,30 @@ void __khugepaged_exit(struct mm_struct *mm)
> > > >  
> > > >  static void release_pte_page(struct page *page)
> > > >  {
> > > > -	dec_node_page_state(page, NR_ISOLATED_ANON + page_is_file_cache(page));
> > > 
> > > I have
> > > 
> > > 	dec_node_page_state(page, NR_ISOLATED_ANON + page_is_file_lru(page));
> > > 
> > > here.  Is there some prerequisite which I wasn't able to find?
> > 
> > The patchset is on top of v5.6. It has been changed since. See
> > 9de4f22a60f7 ("mm: code cleanup for MADV_FREE").
> > 
> > Look like a trivial fixup is required.
> 
> [7/8] makes a big mess.  Can we please have a v4?

Sure. Give me a day.

-- 
 Kirill A. Shutemov