Linux Kernel Selftest development
 help / color / mirror / Atom feed
From: "Ilpo Järvinen" <ilpo.jarvinen@linux.intel.com>
To: Reinette Chatre <reinette.chatre@intel.com>
Cc: fenghua.yu@intel.com, shuah@kernel.org, tony.luck@intel.com,
	peternewman@google.com, babu.moger@amd.com,
	"Maciej Wieczór-Retman" <maciej.wieczor-retman@intel.com>,
	linux-kselftest@vger.kernel.org,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH V3 12/15] selftests/resctrl: Use cache size to determine "fill_buf" buffer size
Date: Fri, 18 Oct 2024 12:06:48 +0300 (EEST)	[thread overview]
Message-ID: <064588b9-f6a9-cf49-bbca-c502314490ba@linux.intel.com> (raw)
In-Reply-To: <388c93423d9a071f5877f3edb4f55ef64bf384bc.1729218182.git.reinette.chatre@intel.com>

[-- Attachment #1: Type: text/plain, Size: 5525 bytes --]

On Thu, 17 Oct 2024, Reinette Chatre wrote:

> By default the MBM and MBA tests use the "fill_buf" benchmark to
> read from a buffer with the goal to measure the memory bandwidth
> generated by this buffer access.
> 
> Care should be taken when sizing the buffer used by the "fill_buf"
> benchmark. If the buffer is small enough to fit in the cache then
> it cannot be expected that the benchmark will generate much memory
> bandwidth. For example, on a system with 320MB L3 cache the existing
> hardcoded default of 250MB is insufficient.
> 
> Use the measured cache size to determine a buffer size that can be
> expected to trigger memory access while keeping the existing default
> as minimum, now renamed to MINIMUM_SPAN, that has been appropriate for
> testing so far.
> 
> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>

Look good, thanks.

Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>

--
 i.

> ---
> Changes since V2:
> - Move duplicate code into helper. (Ilpo)
> - Rename DEFAULT_SPAN to MINIMUM_SPAN to reflect its new purpose. (Ilpo)
> - Do _not_ add Ilpo's Reviewed-by tag ... the patch changed too much.
> 
> Changes since V1:
> - Ensure buffer is at least double L3 cache size. (Ilpo)
> - Support user override of default buffer size. (Ilpo)
> ---
>  tools/testing/selftests/resctrl/fill_buf.c      | 13 +++++++++++++
>  tools/testing/selftests/resctrl/mba_test.c      |  7 ++++++-
>  tools/testing/selftests/resctrl/mbm_test.c      |  7 ++++++-
>  tools/testing/selftests/resctrl/resctrl.h       |  3 ++-
>  tools/testing/selftests/resctrl/resctrl_tests.c |  2 +-
>  5 files changed, 28 insertions(+), 4 deletions(-)
> 
> diff --git a/tools/testing/selftests/resctrl/fill_buf.c b/tools/testing/selftests/resctrl/fill_buf.c
> index 380cc35f10c6..19a01a52dc1a 100644
> --- a/tools/testing/selftests/resctrl/fill_buf.c
> +++ b/tools/testing/selftests/resctrl/fill_buf.c
> @@ -129,3 +129,16 @@ unsigned char *alloc_buffer(size_t buf_size, bool memflush)
>  
>  	return buf;
>  }
> +
> +ssize_t get_fill_buf_size(int cpu_no, const char *cache_type)
> +{
> +	unsigned long cache_total_size = 0;
> +	int ret;
> +
> +	ret = get_cache_size(cpu_no, cache_type, &cache_total_size);
> +	if (ret)
> +		return ret;
> +
> +	return cache_total_size * 2 > MINIMUM_SPAN ?
> +			cache_total_size * 2 : MINIMUM_SPAN;
> +}
> diff --git a/tools/testing/selftests/resctrl/mba_test.c b/tools/testing/selftests/resctrl/mba_test.c
> index 74d95c460bd0..bf37f3555660 100644
> --- a/tools/testing/selftests/resctrl/mba_test.c
> +++ b/tools/testing/selftests/resctrl/mba_test.c
> @@ -182,7 +182,12 @@ static int mba_run_test(const struct resctrl_test *test, const struct user_param
>  		fill_buf.memflush = uparams->fill_buf->memflush;
>  		param.fill_buf = &fill_buf;
>  	} else if (!uparams->benchmark_cmd[0]) {
> -		fill_buf.buf_size = DEFAULT_SPAN;
> +		ssize_t buf_size;
> +
> +		buf_size = get_fill_buf_size(uparams->cpu, "L3");
> +		if (buf_size < 0)
> +			return buf_size;
> +		fill_buf.buf_size = buf_size;
>  		fill_buf.memflush = true;
>  		param.fill_buf = &fill_buf;
>  	}
> diff --git a/tools/testing/selftests/resctrl/mbm_test.c b/tools/testing/selftests/resctrl/mbm_test.c
> index 72261413c868..4224f8ce3538 100644
> --- a/tools/testing/selftests/resctrl/mbm_test.c
> +++ b/tools/testing/selftests/resctrl/mbm_test.c
> @@ -149,7 +149,12 @@ static int mbm_run_test(const struct resctrl_test *test, const struct user_param
>  		fill_buf.memflush = uparams->fill_buf->memflush;
>  		param.fill_buf = &fill_buf;
>  	} else if (!uparams->benchmark_cmd[0]) {
> -		fill_buf.buf_size = DEFAULT_SPAN;
> +		ssize_t buf_size;
> +
> +		buf_size = get_fill_buf_size(uparams->cpu, "L3");
> +		if (buf_size < 0)
> +			return buf_size;
> +		fill_buf.buf_size = buf_size;
>  		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 032cd9ebd761..a553fe975938 100644
> --- a/tools/testing/selftests/resctrl/resctrl.h
> +++ b/tools/testing/selftests/resctrl/resctrl.h
> @@ -41,7 +41,7 @@
>  
>  #define BENCHMARK_ARGS		64
>  
> -#define DEFAULT_SPAN		(250 * MB)
> +#define MINIMUM_SPAN		(250 * MB)
>  
>  /*
>   * fill_buf_param:	"fill_buf" benchmark parameters
> @@ -169,6 +169,7 @@ int perf_event_open(struct perf_event_attr *hw_event, pid_t pid, int cpu,
>  unsigned char *alloc_buffer(size_t buf_size, bool memflush);
>  void mem_flush(unsigned char *buf, size_t buf_size);
>  void fill_cache_read(unsigned char *buf, size_t buf_size, bool once);
> +ssize_t get_fill_buf_size(int cpu_no, const char *cache_type);
>  int initialize_read_mem_bw_imc(void);
>  int measure_read_mem_bw(const struct user_params *uparams,
>  			struct resctrl_val_param *param, pid_t bm_pid);
> diff --git a/tools/testing/selftests/resctrl/resctrl_tests.c b/tools/testing/selftests/resctrl/resctrl_tests.c
> index 88768f4d4961..790f4eb7871a 100644
> --- a/tools/testing/selftests/resctrl/resctrl_tests.c
> +++ b/tools/testing/selftests/resctrl/resctrl_tests.c
> @@ -189,7 +189,7 @@ static struct fill_buf_param *alloc_fill_buf_param(struct user_params *uparams)
>  			ksft_exit_skip("Unable to parse benchmark buffer size.\n");
>  		}
>  	} else {
> -		fill_param->buf_size = DEFAULT_SPAN;
> +		fill_param->buf_size = MINIMUM_SPAN;
>  	}
>  
>  	if (uparams->benchmark_cmd[2]) {
> 

  reply	other threads:[~2024-10-18  9:06 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-18  2:33 [PATCH V3 00/15] selftests/resctrl: Support diverse platforms with MBM and MBA tests Reinette Chatre
2024-10-18  2:33 ` [PATCH V3 01/15] selftests/resctrl: Make functions only used in same file static Reinette Chatre
2024-10-18  2:33 ` [PATCH V3 02/15] selftests/resctrl: Print accurate buffer size as part of MBM results Reinette Chatre
2024-10-18  8:46   ` Ilpo Järvinen
2024-10-18 17:34     ` Reinette Chatre
2024-10-18  2:33 ` [PATCH V3 03/15] selftests/resctrl: Fix memory overflow due to unhandled wraparound Reinette Chatre
2024-10-18  2:33 ` [PATCH V3 04/15] selftests/resctrl: Protect against array overrun during iMC config parsing Reinette Chatre
2024-10-18  2:33 ` [PATCH V3 05/15] selftests/resctrl: Protect against array overflow when reading strings Reinette Chatre
2024-10-18  8:53   ` Ilpo Järvinen
2024-10-18  2:33 ` [PATCH V3 06/15] selftests/resctrl: Make wraparound handling obvious Reinette Chatre
2024-10-18  2:33 ` [PATCH V3 07/15] selftests/resctrl: Remove "once" parameter required to be false Reinette Chatre
2024-10-18  8:54   ` Ilpo Järvinen
2024-10-18  2:33 ` [PATCH V3 08/15] selftests/resctrl: Only support measured read operation Reinette Chatre
2024-10-18  8:55   ` Ilpo Järvinen
2024-10-18  2:33 ` [PATCH V3 09/15] selftests/resctrl: Remove unused measurement code Reinette Chatre
2024-10-18  2:33 ` [PATCH V3 10/15] selftests/resctrl: Make benchmark parameter passing robust Reinette Chatre
2024-10-18  9:03   ` Ilpo Järvinen
2024-10-18 17:34     ` Reinette Chatre
2024-10-18  2:33 ` [PATCH V3 11/15] selftests/resctrl: Ensure measurements skip initialization of default benchmark Reinette Chatre
2024-10-18  2:33 ` [PATCH V3 12/15] selftests/resctrl: Use cache size to determine "fill_buf" buffer size Reinette Chatre
2024-10-18  9:06   ` Ilpo Järvinen [this message]
2024-10-18  2:33 ` [PATCH V3 13/15] selftests/resctrl: Do not compare performance counters and resctrl at low bandwidth Reinette Chatre
2024-10-18  2:33 ` [PATCH V3 14/15] selftests/resctrl: Keep results from first test run Reinette Chatre
2024-10-18  2:33 ` [PATCH V3 15/15] selftests/resctrl: Replace magic constants used as array size Reinette Chatre
2024-10-18  9:08   ` Ilpo Järvinen

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=064588b9-f6a9-cf49-bbca-c502314490ba@linux.intel.com \
    --to=ilpo.jarvinen@linux.intel.com \
    --cc=babu.moger@amd.com \
    --cc=fenghua.yu@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-kselftest@vger.kernel.org \
    --cc=maciej.wieczor-retman@intel.com \
    --cc=peternewman@google.com \
    --cc=reinette.chatre@intel.com \
    --cc=shuah@kernel.org \
    --cc=tony.luck@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