public inbox for linux-arm-kernel@lists.infradead.org
 help / color / mirror / Atom feed
From: David Laight <david.laight.linux@gmail.com>
To: Barry Song <21cnbao@gmail.com>
Cc: Xavier <xavier_qy@163.com>,
	catalin.marinas@arm.com, will@kernel.org,
	akpm@linux-foundation.org, ryan.roberts@arm.com,
	ioworker0@gmail.com, linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH v1] mm/contpte: Optimize loop to reduce redundant operations
Date: Fri, 11 Apr 2025 13:03:09 +0100	[thread overview]
Message-ID: <20250411130309.7894f204@pumpkin> (raw)
In-Reply-To: <CAGsJ_4wsW+urp6gbp+yDG40fZw9Bszny2iJVsV_AzzGqgQy+4g@mail.gmail.com>

On Fri, 11 Apr 2025 09:25:39 +1200
Barry Song <21cnbao@gmail.com> wrote:

> On Mon, Apr 7, 2025 at 9:23 PM Xavier <xavier_qy@163.com> wrote:
> >
> > This commit optimizes the contpte_ptep_get function by adding early
> >  termination logic. It checks if the dirty and young bits of orig_pte
> >  are already set and skips redundant bit-setting operations during
> >  the loop. This reduces unnecessary iterations and improves performance.
> >
> > Signed-off-by: Xavier <xavier_qy@163.com>
> > ---
> >  arch/arm64/mm/contpte.c | 13 +++++++++++--
> >  1 file changed, 11 insertions(+), 2 deletions(-)
> >
> > diff --git a/arch/arm64/mm/contpte.c b/arch/arm64/mm/contpte.c
> > index bcac4f55f9c1..ca15d8f52d14 100644
> > --- a/arch/arm64/mm/contpte.c
> > +++ b/arch/arm64/mm/contpte.c
> > @@ -163,17 +163,26 @@ pte_t contpte_ptep_get(pte_t *ptep, pte_t orig_pte)
> >
> >         pte_t pte;
> >         int i;
> > +       bool dirty = false;
> > +       bool young = false;
> >
> >         ptep = contpte_align_down(ptep);
> >
> >         for (i = 0; i < CONT_PTES; i++, ptep++) {
> >                 pte = __ptep_get(ptep);
> >
> > -               if (pte_dirty(pte))
> > +               if (!dirty && pte_dirty(pte)) {
> > +                       dirty = true;
> >                         orig_pte = pte_mkdirty(orig_pte);
> > +               }
> >
> > -               if (pte_young(pte))
> > +               if (!young && pte_young(pte)) {
> > +                       young = true;
> >                         orig_pte = pte_mkyoung(orig_pte);
> > +               }
> > +
> > +               if (dirty && young)
> > +                       break;  
> 
> This kind of optimization is always tricky. Dev previously tried a similar
> approach to reduce the loop count, but it ended up causing performance
> degradation:
> https://lore.kernel.org/linux-mm/20240913091902.1160520-1-dev.jain@arm.com/
> 
> So we may need actual data to validate this idea.

You might win with 3 loops.
The first looks for both 'dirty' and 'young'.
If it finds only one it jumps to a different loop that continues
the search but only looks for the other flag.

	David

> 
> >         }
> >
> >         return orig_pte;
> > --
> > 2.34.1
> >  
> 
> Thanks
> Barry
> 



  reply	other threads:[~2025-04-11 12:05 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-04-07  9:22 [PATCH v1] mm/contpte: Optimize loop to reduce redundant operations Xavier
2025-04-07 11:29 ` Lance Yang
2025-04-07 12:56   ` Xavier
2025-04-07 13:31     ` Lance Yang
2025-04-07 16:19       ` Dev Jain
2025-04-08  8:58         ` [PATCH v2 0/1] " Xavier
2025-04-08  8:58           ` [PATCH v2 1/1] " Xavier
2025-04-09  4:09             ` Gavin Shan
2025-04-09 15:10               ` Xavier
2025-04-10  0:58                 ` Gavin Shan
2025-04-10  2:53                   ` Xavier
2025-04-10  3:06                     ` Gavin Shan
2025-04-08  9:17         ` [PATCH v1] " Lance Yang
2025-04-09 15:15           ` Xavier
2025-04-10 21:25 ` Barry Song
2025-04-11 12:03   ` David Laight [this message]
2025-04-12  7:18     ` Barry Song
2025-04-11 17:30   ` Dev Jain
2025-04-12  5:05     ` Lance Yang
2025-04-12  5:27       ` Xavier
2025-04-14  8:06       ` Ryan Roberts
2025-04-14  8:51         ` Dev Jain
2025-04-14 12:11           ` Ryan Roberts
2025-04-15  8:22             ` [mm/contpte v3 0/1] " Xavier
2025-04-15  8:22               ` [mm/contpte v3 1/1] " Xavier
2025-04-15  9:01                 ` [mm/contpte v3] " Markus Elfring
2025-04-16  8:57                 ` [mm/contpte v3 1/1] " David Laight
2025-04-16 16:15                   ` Xavier
2025-04-16 12:54                 ` Ryan Roberts
2025-04-16 16:09                   ` Xavier
2025-04-22  9:33                   ` Xavier
2025-04-30 23:17                     ` Barry Song
2025-05-01 12:39                       ` Xavier
2025-05-01 21:19                         ` Barry Song
2025-05-01 21:32                           ` Barry Song
2025-05-04  2:39                             ` Xavier
2025-05-08  1:29                               ` Barry Song
2025-05-08  7:03                             ` [PATCH v4] arm64/mm: Optimize loop to reduce redundant operations of contpte_ptep_get Xavier Xia
2025-05-08  8:30                               ` David Hildenbrand
2025-05-09  9:17                                 ` Xavier
2025-05-09  9:25                                   ` David Hildenbrand
2025-05-09  2:09                               ` Barry Song
2025-05-09  9:20                                 ` Xavier
2025-04-16  2:10               ` [mm/contpte v3 0/1] mm/contpte: Optimize loop to reduce redundant operations Andrew Morton
2025-04-16  3:25                 ` Xavier
2025-04-16 12:47               ` Catalin Marinas
2025-04-16 15:08                 ` Xavier
2025-04-16 12:48               ` Ryan Roberts
2025-04-16 15:22                 ` Xavier

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20250411130309.7894f204@pumpkin \
    --to=david.laight.linux@gmail.com \
    --cc=21cnbao@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=catalin.marinas@arm.com \
    --cc=ioworker0@gmail.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=ryan.roberts@arm.com \
    --cc=will@kernel.org \
    --cc=xavier_qy@163.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox