From: James Houghton <jthoughton@google.com>
To: Sean Christopherson <seanjc@google.com>, kvm@vger.kernel.org
Cc: Maxim Levitsky <mlevitsk@redhat.com>,
Axel Rasmussen <axelrasmussen@google.com>,
Tejun Heo <tj@kernel.org>, Johannes Weiner <hannes@cmpxchg.org>,
mkoutny@suse.com, Yosry Ahmed <yosry.ahmed@linux.dev>,
Yu Zhao <yuzhao@google.com>, David Matlack <dmatlack@google.com>,
James Houghton <jthoughton@google.com>,
cgroups@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH v4 0/7] KVM: selftests: access_tracking_perf_test fixes for NUMA balancing and MGLRU
Date: Thu, 8 May 2025 18:46:41 +0000 [thread overview]
Message-ID: <20250508184649.2576210-1-jthoughton@google.com> (raw)
This series fixes some issues with access_tracking_perf_test when MGLRU
or NUMA balancing are in use.
With MGLRU, touching a page doesn't necessarily clear the Idle flag.
This has come up in the past, and the recommendation was to use MGLRU
generation numbers[1], which this series does.
With NUMA balancing, pages are temporarily mapped as PROT_NONE, so the
SPTEs will be zapped, losing the Accessed bits. The fix here is, in the
event we have lost access information to print a warning and continue
with the test, just like what we do if the test is running a nested VM.
A flag is added for the user to specify if they wish for the test to
always enforce or always skip this check.
Based on kvm/next.
Changelog:
v3[4] -> v4:
- Multiple fixes to access_tracking_perf_test:
1. Fix for software-managed A/D-bit configs.
2. Fix for expected number of pages when guest page size is >
PAGE_SIZE.
3. Properly check youngest generation when looking for pages.
- Do an initial aging pass for MGLRU. This allows us to check that pages
move back to the youngest generation after the first access pass.
- Add controller-root location logic in the cgroup utils, and use this
to make the test work for cgroup-v1 (thanks Sean!).
- Some cleanup for access_tracking_perf_test code, comments, warning
messages, and the commit message (thanks Sean!).
v2[3] -> v3:
- Applied David's directory fix on patch 3.
- Added SoB-by, R-by (patch 2, missed in v2), and A-by.
v1[2] -> v2:
- Re-add clone3_selftests.h for cgroup selftests (thanks Michal!)
- Some comment fixes, patches 2 and 5 (thanks Maxim!).
[1]: https://lore.kernel.org/all/CAOUHufZeADNp_y=Ng+acmMMgnTR=ZGFZ7z-m6O47O=CmJauWjw@mail.gmail.com/
[2]: https://lore.kernel.org/kvm/20250327012350.1135621-1-jthoughton@google.com/
[3]: https://lore.kernel.org/kvm/20250331213025.3602082-1-jthoughton@google.com/
[4]: https://lore.kernel.org/kvm/20250414200929.3098202-1-jthoughton@google.com/
James Houghton (3):
cgroup: selftests: Move cgroup_util into its own library
KVM: selftests: Build and link selftests/cgroup/lib into KVM selftests
KVM: selftests: access_tracking_perf_test: Use MGLRU for access
tracking
Maxim Levitsky (1):
KVM: selftests: access_tracking_perf_test: Add option to skip the
sanity check
Sean Christopherson (3):
KVM: selftests: Extract guts of THP accessor to standalone sysfs
helpers
cgroup: selftests: Move memcontrol specific helpers out of common
cgroup_util.c
cgroup: selftests: Add API to find root of specific controller
tools/testing/selftests/cgroup/Makefile | 21 +-
.../selftests/cgroup/{ => lib}/cgroup_util.c | 118 ++----
.../cgroup/{ => lib/include}/cgroup_util.h | 13 +-
.../testing/selftests/cgroup/lib/libcgroup.mk | 19 +
.../selftests/cgroup/test_memcontrol.c | 78 ++++
tools/testing/selftests/kvm/Makefile.kvm | 4 +-
.../selftests/kvm/access_tracking_perf_test.c | 281 +++++++++++--
.../selftests/kvm/include/lru_gen_util.h | 51 +++
.../testing/selftests/kvm/include/test_util.h | 1 +
.../testing/selftests/kvm/lib/lru_gen_util.c | 386 ++++++++++++++++++
tools/testing/selftests/kvm/lib/test_util.c | 42 +-
11 files changed, 865 insertions(+), 149 deletions(-)
rename tools/testing/selftests/cgroup/{ => lib}/cgroup_util.c (88%)
rename tools/testing/selftests/cgroup/{ => lib/include}/cgroup_util.h (91%)
create mode 100644 tools/testing/selftests/cgroup/lib/libcgroup.mk
create mode 100644 tools/testing/selftests/kvm/include/lru_gen_util.h
create mode 100644 tools/testing/selftests/kvm/lib/lru_gen_util.c
base-commit: 45eb29140e68ffe8e93a5471006858a018480a45
--
2.49.0.1015.ga840276032-goog
next reply other threads:[~2025-05-08 18:46 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-05-08 18:46 James Houghton [this message]
2025-05-08 18:46 ` [PATCH v4 1/7] KVM: selftests: Extract guts of THP accessor to standalone sysfs helpers James Houghton
2025-05-08 18:46 ` [PATCH v4 2/7] KVM: selftests: access_tracking_perf_test: Add option to skip the sanity check James Houghton
2025-05-08 18:46 ` [PATCH v4 3/7] cgroup: selftests: Move memcontrol specific helpers out of common cgroup_util.c James Houghton
2025-05-08 18:46 ` [PATCH v4 4/7] cgroup: selftests: Move cgroup_util into its own library James Houghton
2025-05-08 18:46 ` [PATCH v4 5/7] cgroup: selftests: Add API to find root of specific controller James Houghton
2025-05-08 18:46 ` [PATCH v4 6/7] KVM: selftests: Build and link selftests/cgroup/lib into KVM selftests James Houghton
2025-05-08 18:46 ` [PATCH v4 7/7] KVM: selftests: access_tracking_perf_test: Use MGLRU for access tracking James Houghton
2025-05-16 18:47 ` Sean Christopherson
2025-05-20 16:48 ` [PATCH v4 0/7] KVM: selftests: access_tracking_perf_test fixes for NUMA balancing and MGLRU Sean Christopherson
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=20250508184649.2576210-1-jthoughton@google.com \
--to=jthoughton@google.com \
--cc=axelrasmussen@google.com \
--cc=cgroups@vger.kernel.org \
--cc=dmatlack@google.com \
--cc=hannes@cmpxchg.org \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mkoutny@suse.com \
--cc=mlevitsk@redhat.com \
--cc=seanjc@google.com \
--cc=tj@kernel.org \
--cc=yosry.ahmed@linux.dev \
--cc=yuzhao@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).