From: "Ilpo Järvinen" <ilpo.jarvinen@linux.intel.com>
To: Reinette Chatre <reinette.chatre@intel.com>
Cc: shuah@kernel.org, Dave.Martin@arm.com, james.morse@arm.com,
tony.luck@intel.com, babu.moger@amd.com, fenghuay@nvidia.com,
peternewman@google.com, zide.chen@intel.com,
dapeng1.mi@linux.intel.com, ben.horgan@arm.com,
yu.c.chen@intel.com, jason.zeng@intel.com,
linux-kselftest@vger.kernel.org,
LKML <linux-kernel@vger.kernel.org>,
patches@lists.linux.dev
Subject: Re: [PATCH v3 02/10] selftests/resctrl: Reduce interference from L2 occupancy during cache occupancy test
Date: Thu, 26 Mar 2026 14:56:18 +0200 (EET) [thread overview]
Message-ID: <579d8a41-d6ec-8d9f-93b1-6a6f6312099e@linux.intel.com> (raw)
In-Reply-To: <5a23acd72af3ea0513f730d7773108586f68c430.1773432891.git.reinette.chatre@intel.com>
[-- Attachment #1: Type: text/plain, Size: 6903 bytes --]
On Fri, 13 Mar 2026, Reinette Chatre wrote:
> The CMT test creates a new control group that is also capable of monitoring
> and assigns the workload to it. The workload allocates a buffer that by
> default fills a portion of the L3 and keeps reading from the buffer,
> measuring the L3 occupancy at intervals. The test passes if the workload's
> L3 occupancy is within 15% of the buffer size.
>
> The CMT test does not take into account that some of the workload's data
> may land in L2/L1. Matching L3 occupancy to the size of the buffer while
> a portion of the buffer can be allocated into L2 is not accurate.
>
> Take the L2 cache into account to improve test accuracy:
> - Reduce the workload's L2 cache allocation to the minimum on systems that
> support L2 cache allocation. Do so with a new utility in preparation for
> all L3 cache allocation tests needing the same capability.
> - Increase the buffer size to accommodate data that may be allocated into
> the L2 cache. Use a buffer size double the L3 portion to keep using the
> L3 portion size as goal for L3 occupancy while taking into account that
> some of the data may be in L2.
>
> Running the CMT test on a sample system while introducing significant
> cache misses using "stress-ng --matrix-3d 0 --matrix-3d-zyx" shows
> significant improvement in L3 cache occupancy:
>
> Before:
>
> # Starting CMT test ...
> # Mounting resctrl to "/sys/fs/resctrl"
> # Cache size :335544320
> # Writing benchmark parameters to resctrl FS
> # Write schema "L3:0=fffe0" to resctrl FS
> # Write schema "L3:0=1f" to resctrl FS
> # Benchmark PID: 7089
> # Checking for pass/fail
> # Pass: Check cache miss rate within 15%
> # Percent diff=12
> # Number of bits: 5
> # Average LLC val: 73269248
> # Cache span (bytes): 83886080
> ok 1 CMT: test
>
> After:
> # Starting CMT test ...
> # Mounting resctrl to "/sys/fs/resctrl"
> # Cache size :335544320
> # Writing benchmark parameters to resctrl FS
> # Write schema "L3:0=fffe0" to resctrl FS
> # Write schema "L3:0=1f" to resctrl FS
> # Write schema "L2:1=0x1" to resctrl FS
> # Benchmark PID: 7171
> # Checking for pass/fail
> # Pass: Check cache miss rate within 15%
> # Percent diff=0
> # Number of bits: 5
> # Average LLC val: 83755008
> # Cache span (bytes): 83886080
> ok 1 CMT: test
>
> Reported-by: Dave Martin <Dave.Martin@arm.com>
> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
> Tested-by: Chen Yu <yu.c.chen@intel.com>
> Link: https://lore.kernel.org/lkml/aO+7MeSMV29VdbQs@e133380.arm.com/
> ---
> Changes since v2:
> - New patch split from v1's "selftests/resctrl: Improve accuracy of cache
> occupancy test". (Ilpo)
> - Reword changelog. (Ilpo)
> - Update data used in changelog to match code after patch split.
> - Introduce utility to reduce L2 cache allocation. (Ilpo)
> - Add Chen Yu's tag.
> ---
> tools/testing/selftests/resctrl/cache.c | 13 +++++++++++++
> tools/testing/selftests/resctrl/cmt_test.c | 14 ++++++++++----
> tools/testing/selftests/resctrl/resctrl.h | 3 +++
> 3 files changed, 26 insertions(+), 4 deletions(-)
>
> diff --git a/tools/testing/selftests/resctrl/cache.c b/tools/testing/selftests/resctrl/cache.c
> index 1ff1104e6575..bef71b6feacc 100644
> --- a/tools/testing/selftests/resctrl/cache.c
> +++ b/tools/testing/selftests/resctrl/cache.c
> @@ -173,6 +173,19 @@ int measure_llc_resctrl(const char *filename, pid_t bm_pid)
> return print_results_cache(filename, bm_pid, llc_occu_resc);
> }
>
> +/*
> + * Reduce L2 allocation to minimum when testing L3 cache allocation.
> + */
> +int minimize_l2_occupancy(const struct resctrl_test *test,
> + const struct user_params *uparams,
> + const struct resctrl_val_param *param)
> +{
> + if (!strcmp(test->resource, "L3") && resctrl_resource_exists("L2"))
> + return write_schemata(param->ctrlgrp, "0x1", uparams->cpu, "L2");
> +
> + return 0;
> +}
> +
> /*
> * show_cache_info - Show generic cache test information
> * @no_of_bits: Number of bits
> diff --git a/tools/testing/selftests/resctrl/cmt_test.c b/tools/testing/selftests/resctrl/cmt_test.c
> index 7bc6cf49c1c5..ccb6fe881a94 100644
> --- a/tools/testing/selftests/resctrl/cmt_test.c
> +++ b/tools/testing/selftests/resctrl/cmt_test.c
> @@ -23,7 +23,9 @@
> * Initialize capacity bitmasks (CBMs) of:
> * - control group being tested per test parameters,
> * - default resource group as inverse of control group being tested to prevent
> - * other tasks from interfering with test.
> + * other tasks from interfering with test,
> + * - L2 resource of control group being tested to minimize allocations into
> + * L2 if possible to better predict L3 occupancy.
> */
> static int cmt_init(const struct resctrl_test *test,
> const struct user_params *uparams,
> @@ -46,7 +48,11 @@ static int cmt_init(const struct resctrl_test *test,
> return ret;
>
> snprintf(schemata, sizeof(schemata), "%lx", param->mask);
> - return write_schemata(param->ctrlgrp, schemata, uparams->cpu, test->resource);
> + ret = write_schemata(param->ctrlgrp, schemata, uparams->cpu, test->resource);
> + if (ret)
> + return ret;
> +
> + return minimize_l2_occupancy(test, uparams, param);
> }
>
> static int cmt_setup(const struct resctrl_test *test,
> @@ -175,11 +181,11 @@ static int cmt_run_test(const struct resctrl_test *test, const struct user_param
> span = cache_portion_size(cache_total_size, param.mask, long_mask);
>
> if (uparams->fill_buf) {
> - fill_buf.buf_size = span;
> + fill_buf.buf_size = span * 2;
> fill_buf.memflush = uparams->fill_buf->memflush;
> param.fill_buf = &fill_buf;
> } else if (!uparams->benchmark_cmd[0]) {
> - fill_buf.buf_size = span;
> + fill_buf.buf_size = span * 2;
> fill_buf.memflush = true;
> param.fill_buf = &fill_buf;
> }
> diff --git a/tools/testing/selftests/resctrl/resctrl.h b/tools/testing/selftests/resctrl/resctrl.h
> index c72045c74ac4..7f2ab28be857 100644
> --- a/tools/testing/selftests/resctrl/resctrl.h
> +++ b/tools/testing/selftests/resctrl/resctrl.h
> @@ -216,6 +216,9 @@ int perf_event_reset_enable(int pe_fd);
> int perf_event_measure(int pe_fd, struct perf_event_read *pe_read,
> const char *filename, pid_t bm_pid);
> int measure_llc_resctrl(const char *filename, pid_t bm_pid);
> +int minimize_l2_occupancy(const struct resctrl_test *test,
> + const struct user_params *uparams,
> + const struct resctrl_val_param *param);
> void show_cache_info(int no_of_bits, __u64 avg_llc_val, size_t cache_span, bool lines);
>
> /*
>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
--
i.
next prev parent reply other threads:[~2026-03-26 12:56 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-13 20:32 [PATCH v3 00/10] selftests/resctrl: Fixes and improvements focused on Intel platforms Reinette Chatre
2026-03-13 20:32 ` [PATCH v3 01/10] selftests/resctrl: Improve accuracy of cache occupancy test Reinette Chatre
2026-03-26 12:44 ` Ilpo Järvinen
2026-03-13 20:32 ` [PATCH v3 02/10] selftests/resctrl: Reduce interference from L2 occupancy during " Reinette Chatre
2026-03-26 12:56 ` Ilpo Järvinen [this message]
2026-03-13 20:32 ` [PATCH v3 03/10] selftests/resctrl: Do not store iMC counter value in counter config structure Reinette Chatre
2026-03-13 20:32 ` [PATCH v3 04/10] selftests/resctrl: Prepare for parsing multiple events per iMC Reinette Chatre
2026-03-26 13:03 ` Ilpo Järvinen
2026-03-26 14:34 ` Reinette Chatre
2026-03-13 20:32 ` [PATCH v3 05/10] selftests/resctrl: Support multiple events associated with iMC Reinette Chatre
2026-03-27 17:28 ` Ilpo Järvinen
2026-03-13 20:32 ` [PATCH v3 06/10] selftests/resctrl: Increase size of buffer used in MBM and MBA tests Reinette Chatre
2026-03-27 17:30 ` Ilpo Järvinen
2026-03-13 20:32 ` [PATCH v3 07/10] selftests/resctrl: Raise threshold at which MBM and PMU values are compared Reinette Chatre
2026-03-27 17:34 ` Ilpo Järvinen
2026-03-27 23:19 ` Reinette Chatre
2026-03-13 20:32 ` [PATCH v3 08/10] selftests/resctrl: Remove requirement on cache miss rate Reinette Chatre
2026-03-27 17:45 ` Ilpo Järvinen
2026-03-27 23:21 ` Reinette Chatre
2026-03-31 8:07 ` Ilpo Järvinen
2026-03-31 17:39 ` Reinette Chatre
2026-03-13 20:32 ` [PATCH v3 09/10] selftests/resctrl: Simplify perf usage in CAT test Reinette Chatre
2026-03-27 17:47 ` Ilpo Järvinen
2026-03-13 20:32 ` [PATCH v3 10/10] selftests/resctrl: Reduce L2 impact on " Reinette Chatre
2026-03-27 17:49 ` Ilpo Järvinen
2026-03-27 23:22 ` Reinette Chatre
2026-03-31 19:13 ` [PATCH v3 00/10] selftests/resctrl: Fixes and improvements focused on Intel platforms Shuah Khan
2026-03-31 20:22 ` Reinette Chatre
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=579d8a41-d6ec-8d9f-93b1-6a6f6312099e@linux.intel.com \
--to=ilpo.jarvinen@linux.intel.com \
--cc=Dave.Martin@arm.com \
--cc=babu.moger@amd.com \
--cc=ben.horgan@arm.com \
--cc=dapeng1.mi@linux.intel.com \
--cc=fenghuay@nvidia.com \
--cc=james.morse@arm.com \
--cc=jason.zeng@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=patches@lists.linux.dev \
--cc=peternewman@google.com \
--cc=reinette.chatre@intel.com \
--cc=shuah@kernel.org \
--cc=tony.luck@intel.com \
--cc=yu.c.chen@intel.com \
--cc=zide.chen@intel.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