linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@linux-foundation.org>
To: Yang Shi <shy828301@gmail.com>
Cc: Lance Yang <ioworker0@gmail.com>,
	mhocko@suse.com, zokeefe@google.com, david@redhat.com,
	songmuchun@bytedance.com, peterx@redhat.com, minchan@kernel.org,
	linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 1/1] mm/khugepaged: bypassing unnecessary scans with MMF_DISABLE_THP check
Date: Thu, 22 Feb 2024 13:11:51 -0800	[thread overview]
Message-ID: <20240222131151.4d24563d58240e76a3b535db@linux-foundation.org> (raw)
In-Reply-To: <CAHbLzkoNFn6UE4Hn0gXTm2pLHD7pK4gYUe1zVh3247m30A5u2g@mail.gmail.com>

On Thu, 22 Feb 2024 12:23:21 -0800 Yang Shi <shy828301@gmail.com> wrote:

> > VMA Count |   Old   |   New   |  Change
> > ---------------------------------------
> >     50     |   23us  |    9us  |  -60.9%
> >    100     |   32us  |    9us  |  -71.9%
> >    200     |   44us  |    9us  |  -79.5%
> >    400     |   75us  |    9us  |  -88.0%
> >    800     |   98us  |    9us  |  -90.8%
> >
> > IIUC, once the count of VMAs for the process
> > exceeds page_to_scan, khugepaged needs to
> > wait for scan_sleep_millisecs ms before scanning
> > the next process. IMO, unnecessary scans could
> > actually be skipped with a very inexpensive
> > mm->flags check in this case.
> 
> Thanks for following up on this, can you please capture all the
> information in the commit log?

I added it.

--- a/txt/mm-khugepaged-bypassing-unnecessary-scans-with-mmf_disable_thp-check.txt
+++ b/txt/mm-khugepaged-bypassing-unnecessary-scans-with-mmf_disable_thp-check.txt
@@ -9,6 +9,24 @@ and the MMF_DISABLE_THP flag is set later, this scanning process
 becomes unnecessary for that mm and can be skipped to avoid redundant
 operations, especially in scenarios with a large address space.
 
+On an Intel Core i5 CPU, the time taken by khugepaged to scan the
+address space of the process, which has been set with the
+MMF_DISABLE_THP flag after being added to the mm_slots list, is as
+follows (shorter is better):
+
+VMA Count |   Old   |   New   |  Change
+---------------------------------------
+    50    |   23us  |    9us  |  -60.9%
+   100    |   32us  |    9us  |  -71.9%
+   200    |   44us  |    9us  |  -79.5%
+   400    |   75us  |    9us  |  -88.0%
+   800    |   98us  |    9us  |  -90.8%
+
+Once the count of VMAs for the process exceeds page_to_scan, khugepaged
+needs to wait for scan_sleep_millisecs ms before scanning the next
+process.  IMO, unnecessary scans could actually be skipped with a very
+inexpensive mm->flags check in this case.
+
 This commit introduces a check before each scanning process to test the
 MMF_DISABLE_THP flag for the given mm; if the flag is set, the scanning
 process is bypassed, thereby improving the efficiency of khugepaged.


  reply	other threads:[~2024-02-22 21:11 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-01-29  5:45 [PATCH 1/1] mm/khugepaged: bypassing unnecessary scans with MMF_DISABLE_THP check Lance Yang
2024-01-29 16:28 ` Michal Hocko
2024-01-30  2:12   ` Lance Yang
2024-01-30  3:08     ` Lance Yang
2024-01-30  9:35       ` Michal Hocko
2024-01-30  9:46         ` Lance Yang
2024-01-31  0:37           ` Lance Yang
2024-01-29 18:53 ` Yang Shi
2024-01-29 19:03   ` Zach O'Keefe
2024-01-30  2:37     ` Lance Yang
2024-01-30  2:21   ` Lance Yang
2024-01-31  9:30 ` Lance Yang
2024-01-31 20:06   ` Yang Shi
2024-02-01  1:13     ` Lance Yang
2024-02-01 18:56       ` Yang Shi
2024-02-03  3:20         ` Lance Yang
2024-02-21 22:11   ` Andrew Morton
2024-02-22  7:43     ` Lance Yang
2024-02-22  7:51   ` Lance Yang
2024-02-22  8:51     ` David Hildenbrand
2024-02-22  9:12       ` Lance Yang
2024-02-22 20:23     ` Yang Shi
2024-02-22 21:11       ` Andrew Morton [this message]
2024-02-23  7:59         ` Lance Yang
2024-02-24  1:47         ` Yang Shi
2024-02-24  1:46 ` Yang Shi
2024-02-24  3:54   ` Lance Yang
2024-02-25  4:56     ` Lance Yang

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=20240222131151.4d24563d58240e76a3b535db@linux-foundation.org \
    --to=akpm@linux-foundation.org \
    --cc=david@redhat.com \
    --cc=ioworker0@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@suse.com \
    --cc=minchan@kernel.org \
    --cc=peterx@redhat.com \
    --cc=shy828301@gmail.com \
    --cc=songmuchun@bytedance.com \
    --cc=zokeefe@google.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;
as well as URLs for NNTP newsgroup(s).