From: Mauricio Faria de Oliveira <mfo@igalia.com>
To: Andrew Morton <akpm@linux-foundation.org>,
Vlastimil Babka <vbabka@suse.cz>
Cc: Oscar Salvador <osalvador@suse.de>,
Suren Baghdasaryan <surenb@google.com>,
Michal Hocko <mhocko@suse.com>,
Brendan Jackman <jackmanb@google.com>,
Johannes Weiner <hannes@cmpxchg.org>, Zi Yan <ziy@nvidia.com>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
kernel-dev@igalia.com
Subject: [PATCH 0/3] mm/page_owner: add options 'print_handle' and 'print_stack' for 'show_stacks'
Date: Wed, 24 Sep 2025 14:40:20 -0300 [thread overview]
Message-ID: <20250924174023.261125-1-mfo@igalia.com> (raw)
Problem:
The use case of monitoring the memory usage per stack trace (or tracking
a particular stack trace) requires uniquely identifying each stack trace.
This has to be done for every stack trace on every sample of monitoring,
even if tracking only one stack trace (to identify it among all others).
Therefore, an approach like, for example, hashing the stack traces from
'show_stacks' for calculating unique identifiers can become expensive.
Solution:
Fortunately, the kernel can provide a unique identifier for stack traces
in page_owner, which is the handle number in stackdepot.
Additionally, with that information, the stack traces themselves are not
needed until the time when the memory usage should be associated with a
stack trace (say, to look at a few top consumers), using handle numbers.
This eliminates hashing and reduces filtering related to stack traces in
userspace, and reduces text emitted/copied by the kernel.
Changes:
This patchset adds 2 options to configure the output of 'show_stacks':
- print_handle: print the handle number of the stack traces (disabled)
- print_stack: print the stack traces (enabled)
Now, it is possible to sample with handles/without stacks, and do one last
sample with handles and stacks to associate the handles with their stacks.
Tested on next-20250922.
Results:
Numbers on a freshly booted system (gains should be higher after load,
as more stack traces would be present); average time of 10 runs:
- Baseline:
average = 0.165
# cat /sys/kernel/debug/page_owner_stacks/show_stacks >file
- Enable handles:
average = 0.165 (same)
# echo 1 >/sys/kernel/debug/page_owner_stacks/print_handle
# cat /sys/kernel/debug/page_owner_stacks/show_stacks >file
- Enable handles, remove stack traces in userspace (example):
average = 0.167 (+1.2%)
# echo 1 >/sys/kernel/debug/page_owner_stacks/print_handle
# grep -v '^ ' /sys/kernel/debug/page_owner_stacks/show_stacks >file
- Enable handles, disable stack traces in kernel:
average = 0.041 (-75.2%)
# echo 1 >/sys/kernel/debug/page_owner_stacks/print_handle
# echo 0 >/sys/kernel/debug/page_owner_stacks/print_stack
# cat /sys/kernel/debug/page_owner_stacks/show_stacks >file
Mauricio Faria de Oliveira (3):
mm/page_owner: add option 'print_handle' for 'show_stacks'
mm/page_owner: add option 'print_stack' for 'show_stacks'
mm/page_owner: update Documentation with print_handle and print_stack
Documentation/mm/page_owner.rst | 15 ++++++++++++++-
mm/page_owner.c | 18 ++++++++++++++----
2 files changed, 28 insertions(+), 5 deletions(-)
--
2.48.1
next reply other threads:[~2025-09-24 17:40 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-09-24 17:40 Mauricio Faria de Oliveira [this message]
2025-09-24 17:40 ` [PATCH 1/3] mm/page_owner: add option 'print_handle' " Mauricio Faria de Oliveira
2025-09-25 20:28 ` Joshua Hahn
2025-09-25 22:25 ` Mauricio Faria de Oliveira
2025-09-24 17:40 ` [PATCH 2/3] mm/page_owner: add option 'print_stack' " Mauricio Faria de Oliveira
2025-09-24 17:40 ` [PATCH 3/3] mm/page_owner: update Documentation with 'print_handle' and 'print_stack' Mauricio Faria de Oliveira
2025-09-25 16:08 ` [PATCH 0/3] mm/page_owner: add options 'print_handle' and 'print_stack' for 'show_stacks' Michal Hocko
2025-09-25 19:38 ` Mauricio Faria de Oliveira
2025-09-26 6:55 ` Michal Hocko
2025-09-26 16:47 ` Mauricio Faria de Oliveira
2025-09-30 14:02 ` Michal Hocko
2025-09-30 14:32 ` Mauricio Faria de Oliveira
2025-10-01 10:58 ` Vlastimil Babka
2025-10-01 17:37 ` Mauricio Faria de Oliveira
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=20250924174023.261125-1-mfo@igalia.com \
--to=mfo@igalia.com \
--cc=akpm@linux-foundation.org \
--cc=hannes@cmpxchg.org \
--cc=jackmanb@google.com \
--cc=kernel-dev@igalia.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.com \
--cc=osalvador@suse.de \
--cc=surenb@google.com \
--cc=vbabka@suse.cz \
--cc=ziy@nvidia.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
all inboxes | Powered by JetHome®