public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Reinette Chatre <reinette.chatre@intel.com>
To: "Ilpo Järvinen" <ilpo.jarvinen@linux.intel.com>,
	linux-kselftest@vger.kernel.org, "Shuah Khan" <shuah@kernel.org>,
	"Shaopeng Tan" <tan.shaopeng@jp.fujitsu.com>,
	"Maciej Wieczór-Retman" <maciej.wieczor-retman@intel.com>,
	"Fenghua Yu" <fenghua.yu@intel.com>
Cc: <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 06/24] selftests/resctrl: Exclude shareable bits from schemata in CAT test
Date: Thu, 2 Nov 2023 10:48:01 -0700	[thread overview]
Message-ID: <83479a22-ef2c-4577-aa07-d4cc2cf90fe6@intel.com> (raw)
In-Reply-To: <20231024092634.7122-7-ilpo.jarvinen@linux.intel.com>

Hi Ilpo,

On 10/24/2023 2:26 AM, Ilpo Järvinen wrote:
> CAT test doesn't take shareable bits into account, i.e., the test might
> be sharing cache with some devices (e.g., graphics).
> 
> Introduce get_mask_no_shareable() and use it to provision an
> environment for CAT test where the allocated LLC is isolated better.
> Excluding shareable_bits may create hole(s) into the cbm_mask, thus add
> a new helper count_contiguous_bits() to find the longest contiguous set
> of CBM bits.
> 
> create_bit_mask() is needed by an upcoming CAT test rewrite so make it
> available in resctrl.h right away.
> 
> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
> ---
>  tools/testing/selftests/resctrl/cat_test.c  | 12 ++-
>  tools/testing/selftests/resctrl/resctrl.h   |  3 +
>  tools/testing/selftests/resctrl/resctrlfs.c | 84 +++++++++++++++++++++
>  3 files changed, 95 insertions(+), 4 deletions(-)
> 
> diff --git a/tools/testing/selftests/resctrl/cat_test.c b/tools/testing/selftests/resctrl/cat_test.c
> index 80861c362a53..e5861e7cba7e 100644
> --- a/tools/testing/selftests/resctrl/cat_test.c
> +++ b/tools/testing/selftests/resctrl/cat_test.c
> @@ -92,13 +92,17 @@ int cat_perf_miss_val(int cpu_no, int n, char *cache_type)
>  	unsigned long l_mask, l_mask_1;
>  	int ret, pipefd[2], sibling_cpu_no;
>  	unsigned long cache_total_size = 0;
> -	unsigned long long_mask;
> +	unsigned long full_cache_mask, long_mask;

Please do maintain reverse fir order.

>  	int count_of_bits;
>  	char pipe_message;
>  	size_t span;
>  
>  	/* Get default cbm mask for L3/L2 cache */
> -	ret = get_cbm_mask(cache_type, &long_mask);
> +	ret = get_cbm_mask(cache_type, &full_cache_mask);
> +	if (ret)
> +		return ret;
> +	/* Get the exclusive portion of the cache */
> +	ret = get_mask_no_shareable(cache_type, &long_mask);
>  	if (ret)
>  		return ret;
>  
> +
> +/*
> + * count_contiguous_bits - Returns the longest train of bits in a bit mask
> + * @val		A bit mask
> + * @start	The location of the least-significant bit of the longest train
> + *
> + * Return:	The length of the contiguous bits in the longest train of bits
> + */
> +static unsigned int count_contiguous_bits(unsigned long val, unsigned int *start)
> +{
> +	unsigned long last_val;
> +	int count = 0;

could count be unsigned int?

> +
> +	while (val) {
> +		last_val = val;
> +		val &= (val >> 1);
> +		count++;
> +	}
> +
> +	if (start) {
> +		if (count)
> +			*start = ffsl(last_val) - 1;
> +		else
> +			*start = 0;
> +	}
> +
> +	return count;
> +}
> +
>  /*
>   * get_cbm_mask - Get cbm bit mask
>   * @cache_type:		Cache level L2/L3
> @@ -253,6 +291,52 @@ int get_cbm_mask(const char *cache_type, unsigned long *mask)
>  	return 0;
>  }
>  
> +/*
> + * get_shareable_mask - Get shareable mask from shareable_bits for given cache
> + * @cache_type:		Cache level L2/L3
> + * @shareable_mask:	shareable mask returned as unsigned long

A nitpick but please be consistent in starting sentences with capital letter.

> + *
> + * Return: = 0 on success, < 0 on failure.
> + */
> +int get_shareable_mask(const char *cache_type, unsigned long *shareable_mask)
> +{
> +	char mask_path[1024];
> +
> +	if (!cache_type)
> +		return -1;

Same question as earlier about error checking.

> +
> +	snprintf(mask_path, sizeof(mask_path), "%s/%s/shareable_bits",
> +		 INFO_PATH, cache_type);
> +
> +	return get_bit_mask(mask_path, shareable_mask);
> +}
> +
> +/*
> + * get_mask_no_shareable - Get CBM mask without shareable_bits for given cache
> + * @cache_type:		Cache level L2/L3
> + * @mask:		mask returned as unsigned long

Apart from comment about capital letter this comment does not seem to
reveal more about what code does.

> + *
> + * Return: = 0 on success, < 0 on failure.
> + */
> +int get_mask_no_shareable(const char *cache_type, unsigned long *mask)
> +{
> +	unsigned long full_mask, shareable_mask;
> +	unsigned int start, len;
> +
> +	if (get_cbm_mask(cache_type, &full_mask) < 0)
> +		return -1;
> +	if (get_shareable_mask(cache_type, &shareable_mask) < 0)
> +		return -1;
> +
> +	len = count_contiguous_bits(full_mask & ~shareable_mask, &start);
> +	if (!len)
> +		return -1;
> +
> +	*mask = create_bit_mask(start, len);
> +
> +	return 0;
> +}

Nicely done, thanks.

> +
>  /*
>   * get_core_sibling - Get sibling core id from the same socket for given CPU
>   * @cpu_no:	CPU number


Reinette

  reply	other threads:[~2023-11-02 17:48 UTC|newest]

Thread overview: 72+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-10-24  9:26 [PATCH 00/24] selftests/resctrl: CAT test improvements & generalized test framework Ilpo Järvinen
2023-10-24  9:26 ` [PATCH 01/24] selftests/resctrl: Split fill_buf to allow tests finer-grained control Ilpo Järvinen
2023-10-27 11:24   ` Maciej Wieczór-Retman
2023-11-02 17:46   ` Reinette Chatre
2023-10-24  9:26 ` [PATCH 02/24] selftests/resctrl: Refactor fill_buf functions Ilpo Järvinen
2023-10-27 11:32   ` Maciej Wieczór-Retman
2023-10-27 11:41     ` Ilpo Järvinen
2023-10-24  9:26 ` [PATCH 03/24] selftests/resctrl: Refactor get_cbm_mask() Ilpo Järvinen
2023-10-27 11:39   ` Maciej Wieczór-Retman
2023-11-02 17:47   ` Reinette Chatre
2023-11-03 12:09     ` Ilpo Järvinen
2023-10-24  9:26 ` [PATCH 04/24] selftests/resctrl: Mark get_cache_size() cache_type const Ilpo Järvinen
2023-10-24  9:26 ` [PATCH 05/24] selftests/resctrl: Create cache_size() helper Ilpo Järvinen
2023-11-02 17:47   ` Reinette Chatre
2023-11-03  8:53     ` Ilpo Järvinen
2023-11-03 22:49       ` Reinette Chatre
2023-10-24  9:26 ` [PATCH 06/24] selftests/resctrl: Exclude shareable bits from schemata in CAT test Ilpo Järvinen
2023-11-02 17:48   ` Reinette Chatre [this message]
2023-10-24  9:26 ` [PATCH 07/24] selftests/resctrl: Split measure_cache_vals() function Ilpo Järvinen
2023-11-02 17:48   ` Reinette Chatre
2023-10-24  9:26 ` [PATCH 08/24] selftests/resctrl: Split show_cache_info() to test specific and generic parts Ilpo Järvinen
2023-11-02 17:48   ` Reinette Chatre
2023-10-24  9:26 ` [PATCH 09/24] selftests/resctrl: Remove unnecessary __u64 -> unsigned long conversion Ilpo Järvinen
2023-11-02 17:48   ` Reinette Chatre
2023-11-03  9:19     ` Ilpo Järvinen
2023-10-24  9:26 ` [PATCH 10/24] selftests/resctrl: Remove nested calls in perf event handling Ilpo Järvinen
2023-11-02 17:49   ` Reinette Chatre
2023-11-03  9:45     ` Ilpo Järvinen
2023-10-24  9:26 ` [PATCH 11/24] selftests/resctrl: Consolidate naming of perf event related things Ilpo Järvinen
2023-10-24  9:26 ` [PATCH 12/24] selftests/resctrl: Improve perf init Ilpo Järvinen
2023-10-27 11:45   ` Maciej Wieczór-Retman
2023-10-24  9:26 ` [PATCH 13/24] selftests/resctrl: Convert perf related globals to locals Ilpo Järvinen
2023-10-27 11:47   ` Maciej Wieczór-Retman
2023-10-24  9:26 ` [PATCH 14/24] selftests/resctrl: Move cat_val() to cat_test.c and rename to cat_test() Ilpo Järvinen
2023-10-27 11:51   ` Maciej Wieczór-Retman
2023-10-27 12:18     ` Ilpo Järvinen
2023-10-24  9:26 ` [PATCH 15/24] selftests/resctrl: Read in less obvious order to defeat prefetch optimizations Ilpo Järvinen
2023-10-24  9:26 ` [PATCH 16/24] selftests/resctrl: Rewrite Cache Allocation Technology (CAT) test Ilpo Järvinen
2023-10-27 12:05   ` Maciej Wieczór-Retman
     [not found]     ` <fb5e1a50-ba7-1ee8-8bf2-bb8b64b27b1@linux.intel.com>
2023-10-31  7:24       ` Maciej Wieczór-Retman
2023-11-02 17:51   ` Reinette Chatre
2023-11-03 10:57     ` Ilpo Järvinen
2023-11-03 22:50       ` Reinette Chatre
2023-10-24  9:26 ` [PATCH 17/24] selftests/resctrl: Create struct for input parameter Ilpo Järvinen
2023-10-27 12:07   ` Maciej Wieczór-Retman
2023-11-02 17:51   ` Reinette Chatre
2023-11-03 11:24     ` Ilpo Järvinen
2023-11-03 22:50       ` Reinette Chatre
2023-11-06  9:06         ` Ilpo Järvinen
2023-10-24  9:26 ` [PATCH 18/24] selftests/resctrl: Introduce generalized test framework Ilpo Järvinen
2023-11-02 17:52   ` Reinette Chatre
2023-11-03  9:54     ` Ilpo Järvinen
2023-11-03 22:50       ` Reinette Chatre
2023-10-24  9:26 ` [PATCH 19/24] selftests/resctrl: Pass write_schemata() resource instead of test name Ilpo Järvinen
2023-10-24  9:26 ` [PATCH 20/24] selftests/resctrl: Add helper to convert L2/3 to integer Ilpo Järvinen
2023-10-27 12:09   ` Maciej Wieczór-Retman
2023-10-24  9:26 ` [PATCH 21/24] selftests/resctrl: Get resource id from cache id Ilpo Järvinen
     [not found]   ` <cb2ctfignowlom7lb2t5zhdgtm4s2jlzlvtumlnvxecwwtjk34@ysgepmgkv6bb>
     [not found]     ` <ab4c6aa5-ea49-363a-ff7b-2215665f185d@linux.intel.com>
2023-10-31  7:58       ` Maciej Wieczór-Retman
2023-10-24  9:26 ` [PATCH 22/24] selftests/resctrl: Add test groups and name L3 CAT test L3_CAT Ilpo Järvinen
2023-10-24  9:26 ` [PATCH 23/24] selftests/resctrl: Add L2 CAT test Ilpo Järvinen
2023-11-02 17:57   ` Reinette Chatre
2023-11-03 10:39     ` Ilpo Järvinen
2023-11-03 22:53       ` Reinette Chatre
2023-11-06  9:53         ` Ilpo Järvinen
2023-11-06 17:03           ` Reinette Chatre
2023-11-06 21:22             ` Reinette Chatre
2023-11-07  9:33               ` Ilpo Järvinen
2023-11-08 16:31                 ` Reinette Chatre
2023-10-24  9:26 ` [PATCH 24/24] selftests/resctrl: Ignore failures from L2 CAT test with <= 2 bits Ilpo Järvinen
2023-11-02 17:57   ` Reinette Chatre
2023-11-03 10:24     ` Ilpo Järvinen
2023-11-03 22:53       ` 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=83479a22-ef2c-4577-aa07-d4cc2cf90fe6@intel.com \
    --to=reinette.chatre@intel.com \
    --cc=fenghua.yu@intel.com \
    --cc=ilpo.jarvinen@linux.intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-kselftest@vger.kernel.org \
    --cc=maciej.wieczor-retman@intel.com \
    --cc=shuah@kernel.org \
    --cc=tan.shaopeng@jp.fujitsu.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