linux-doc.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Moger, Babu" <bmoger@amd.com>
To: Reinette Chatre <reinette.chatre@intel.com>,
	babu.moger@amd.com, corbet@lwn.net, tglx@linutronix.de,
	mingo@redhat.com, bp@alien8.de, dave.hansen@linux.intel.com
Cc: fenghua.yu@intel.com, x86@kernel.org, hpa@zytor.com,
	thuth@redhat.com, paulmck@kernel.org, rostedt@goodmis.org,
	akpm@linux-foundation.org, xiongwei.song@windriver.com,
	pawan.kumar.gupta@linux.intel.com,
	daniel.sneddon@linux.intel.com, perry.yuan@amd.com,
	sandipan.das@amd.com, kai.huang@intel.com, xiaoyao.li@intel.com,
	seanjc@google.com, jithu.joseph@intel.com, brijesh.singh@amd.com,
	xin3.li@intel.com, ebiggers@google.com,
	andrew.cooper3@citrix.com, mario.limonciello@amd.com,
	james.morse@arm.com, tan.shaopeng@fujitsu.com,
	tony.luck@intel.com, linux-doc@vger.kernel.org,
	linux-kernel@vger.kernel.org, peternewman@google.com,
	maciej.wieczor-retman@intel.com, eranian@google.com,
	jpoimboe@kernel.org, thomas.lendacky@amd.com
Subject: Re: [PATCH v9 08/26] x86/resctrl: Introduce the interface to display monitor mode
Date: Fri, 22 Nov 2024 12:25:08 -0600	[thread overview]
Message-ID: <0eff9462-e7e2-49a9-9538-c8907024322f@amd.com> (raw)
In-Reply-To: <d5bead60-6a38-426e-afa6-e7980a268d6f@intel.com>

Hi Reinette,

On 11/18/2024 4:07 PM, Reinette Chatre wrote:
> Hi Babu,
> 
> On 11/18/24 11:04 AM, Moger, Babu wrote:
>> Hi Reinette,
>>
>> On 11/15/24 18:00, Reinette Chatre wrote:
>>> Hi Babu,
>>>
>>> On 10/29/24 4:21 PM, Babu Moger wrote:
>>>> Introduce the interface file "mbm_assign_mode" to list monitor modes
>>>> supported.
>>>>
>>>> The "mbm_cntr_assign" mode provides the option to assign a counter to
>>>> an RMID, event pair and monitor the bandwidth as long as it is assigned.
>>>>
>>>> On AMD systems "mbm_cntr_assign" is backed by the ABMC (Assignable
>>>> Bandwidth Monitoring Counters) hardware feature and is enabled by default.
>>>>
>>>> The "default" mode is the existing monitoring mode that works without the
>>>> explicit counter assignment, instead relying on dynamic counter assignment
>>>> by hardware that may result in hardware not dedicating a counter resulting
>>>> in monitoring data reads returning "Unavailable".
>>>>
>>>> Provide an interface to display the monitor mode on the system.
>>>> $ cat /sys/fs/resctrl/info/L3_MON/mbm_assign_mode
>>>> [mbm_cntr_assign]
>>>> default
>>>>
>>>> Signed-off-by: Babu Moger <babu.moger@amd.com>
>>>> ---
> 
> ...
> 
>>> I'm concerned that users with Intel platforms may want to use the "mbm_cntr_assign" mode
>>> to make the event data "more predictable" and then be concerned when the mode does
>>> not exist.
>>>
>>> As an alternative, is it possible to know the number of hardware counters on AMD systems
>>> without ABMC? I wonder if we could perhaps always expose num_mbm_cntrs as a way for
>>> users to know if their platform may be impacted by this type of "unpredictability" (by comparing
>>> num_mbm_cntrs to num_rmids).
>>
>> There is some round about(or hacky) way to find that out number of RMIDs
>> that can be active.
> 
> Does this give consistent and accurate data? Is this something that can be added to resctrl?
> (Reading your other message [1] it does not sound as though it can produce an accurate
> number on boot.)
> If not then it will be up to the documentation to be accurate.
> 
> 
>>>> +
>>>> +	AMD Platforms with ABMC (Assignable Bandwidth Monitoring Counters) feature
>>>> +	enable this mode by default so that counters remain assigned even when the
>>>> +	corresponding RMID is not in use by any processor.
>>>> +
>>>> +	"default":
>>>> +
>>>> +	In default mode resctrl assumes there is a hardware counter for each
>>>> +	event within every CTRL_MON and MON group. Reading mbm_total_bytes or
>>>> +	mbm_local_bytes may report 'Unavailable' if there is no counter associated
>>>> +	with that event.
>>>
>>> If I understand correctly, on AMD platforms without ABMC the events only report
>>> "Unavailable" if there is no counter assigned at the time of the query. If a counter
>>> is unassigned and then reassigned then the event count will reset and the user
>>> will get some data back but it may thus be unpredictable (to match earlier language).
>>> Is this correct? Any AMD platform in "default" mode may thus be vulnerable to
>>> "unpredictable" event counts (not just "Unavailable") ... this gets complicated
>>
>> Yes. All the AMD systems without ABMC are affected by this problem.
>>
>>> because users should be steered to avoid "default" mode if mbm_assign_mode is
>>> available, while not be made concerned to use "default" mode on Intel where
>>> mbm_assign_mode is not available.
>>
>> Can we add text to clarify this?
> 
> Please do.

I think we need to add text about AMD systems. How about this?

"default":
In default mode resctrl assumes there is a hardware counter for each
event within every CTRL_MON and MON group. On AMD systems with 16 more 
monitoring groups, reading mbm_total_bytes or mbm_local_bytes may report 
'Unavailable' if there is no counter associated with that event. It is 
therefore recommended to use the 'mbm_cntr_assign' mode, if supported."

> 
> Reinette
> 
> [1] https://lore.kernel.org/all/35fc70fd-0281-4ac8-b32b-efa2f4516901@amd.com/
> 

-- 
- Babu Moger


  reply	other threads:[~2024-11-22 18:25 UTC|newest]

Thread overview: 115+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-29 23:21 [PATCH v9 00/26] x86/resctrl : Support AMD Assignable Bandwidth Monitoring Counters (ABMC) Babu Moger
2024-10-29 23:21 ` [PATCH v9 01/26] x86/resctrl: Add __init attribute for the functions called in resctrl_late_init Babu Moger
2024-11-15 23:21   ` Reinette Chatre
2024-11-18 17:44     ` Moger, Babu
2024-11-18 22:07       ` Reinette Chatre
2024-11-20 20:02         ` Moger, Babu
2024-10-29 23:21 ` [PATCH v9 02/26] x86/cpufeatures: Add support for Assignable Bandwidth Monitoring Counters (ABMC) Babu Moger
2024-10-29 23:21 ` [PATCH v9 03/26] x86/resctrl: Add ABMC feature in the command line options Babu Moger
2024-10-29 23:21 ` [PATCH v9 04/26] x86/resctrl: Consolidate monitoring related data from rdt_resource Babu Moger
2024-10-29 23:21 ` [PATCH v9 05/26] x86/resctrl: Detect Assignable Bandwidth Monitoring feature details Babu Moger
2024-10-29 23:21 ` [PATCH v9 06/26] x86/resctrl: Introduce resctrl_file_fflags_init() to initialize fflags Babu Moger
2024-10-29 23:21 ` [PATCH v9 07/26] x86/resctrl: Add support to enable/disable AMD ABMC feature Babu Moger
2024-10-29 23:21 ` [PATCH v9 08/26] x86/resctrl: Introduce the interface to display monitor mode Babu Moger
2024-11-16  0:00   ` Reinette Chatre
2024-11-18 19:04     ` Moger, Babu
2024-11-18 22:07       ` Reinette Chatre
2024-11-22 18:25         ` Moger, Babu [this message]
2024-11-22 21:37           ` Reinette Chatre
2024-11-23  0:02             ` Moger, Babu
2024-11-25 18:17               ` Reinette Chatre
2024-11-26 17:09                 ` Moger, Babu
2024-11-26 19:01                   ` Reinette Chatre
2024-11-26 21:57                     ` Moger, Babu
2024-10-29 23:21 ` [PATCH v9 09/26] x86/resctrl: Introduce interface to display number of monitoring counters Babu Moger
2024-11-16  0:06   ` Reinette Chatre
2024-11-18 21:31     ` Moger, Babu
2025-02-03 13:26       ` Peter Newman
2024-10-29 23:21 ` [PATCH v9 10/26] x86/resctrl: Introduce bitmap mbm_cntr_free_map to track assignable counters Babu Moger
2024-11-16  0:11   ` Reinette Chatre
2024-10-29 23:21 ` [PATCH v9 11/26] x86/resctrl: Introduce mbm_total_cfg and mbm_local_cfg in struct rdt_hw_mon_domain Babu Moger
2024-10-29 23:21 ` [PATCH v9 12/26] x86/resctrl: Remove MSR reading of event configuration value Babu Moger
2024-11-16  0:24   ` Reinette Chatre
2024-11-19 16:50     ` Moger, Babu
2024-10-29 23:21 ` [PATCH v9 13/26] x86/resctrl: Introduce mbm_cntr_map to track assignable counters at domain Babu Moger
2024-10-29 23:21 ` [PATCH v9 14/26] x86/resctrl: Introduce interface to display number of free counters Babu Moger
2024-10-29 23:57   ` Luck, Tony
2024-10-30 14:15     ` Moger, Babu
2024-11-04 14:14   ` Peter Newman
2024-11-04 17:31     ` Moger, Babu
2024-11-16  0:31   ` Reinette Chatre
2024-11-19 19:20     ` Moger, Babu
2024-11-21 21:12       ` Reinette Chatre
2024-11-22 23:36         ` Moger, Babu
2024-11-25 19:00           ` Reinette Chatre
2024-11-26 23:31             ` Moger, Babu
2024-11-26 23:56               ` Reinette Chatre
2024-11-27 14:57                 ` Moger, Babu
2024-11-27 19:05                   ` Reinette Chatre
2024-11-28 11:10                     ` Peter Newman
2024-11-28 19:35                       ` Moger, Babu
2024-11-29  9:59                         ` Peter Newman
2024-11-29 17:06                           ` Moger, Babu
2024-12-02 10:43                             ` Peter Newman
2024-12-02 15:02                               ` Moger, Babu
2024-12-02 18:33                             ` Reinette Chatre
2024-12-02 19:48                               ` Moger, Babu
2024-12-02 20:15                                 ` Reinette Chatre
2024-12-02 20:42                                   ` Moger, Babu
2024-12-02 21:09                                     ` Reinette Chatre
2024-12-02 21:28                                       ` Moger, Babu
2024-12-02 21:47                                         ` Reinette Chatre
2024-12-02 22:06                                           ` Moger, Babu
2024-10-29 23:21 ` [PATCH v9 15/26] x86/resctrl: Add data structures and definitions for ABMC assignment Babu Moger
2024-11-16  0:35   ` Reinette Chatre
2024-10-29 23:21 ` [PATCH v9 16/26] x86/resctrl: Introduce cntr_id in mongroup for assignments Babu Moger
2024-11-16  0:38   ` Reinette Chatre
2024-11-19 20:02     ` Moger, Babu
2024-10-29 23:21 ` [PATCH v9 17/26] x86/resctrl: Implement resctrl_arch_config_cntr() to assign a counter with ABMC Babu Moger
2024-10-29 23:54   ` Luck, Tony
2024-10-30 14:14     ` Moger, Babu
2024-11-16  0:44   ` Reinette Chatre
2024-11-19 20:12     ` Moger, Babu
2024-11-21 20:18       ` Reinette Chatre
2024-11-22 18:54         ` Moger, Babu
2024-11-22 21:52           ` Reinette Chatre
2024-11-23  0:15             ` Moger, Babu
2024-10-29 23:21 ` [PATCH v9 18/26] x86/resctrl: Add the interface to assign/update counter assignment Babu Moger
2024-11-16  0:57   ` Reinette Chatre
2024-11-20 18:05     ` Moger, Babu
2024-11-21 20:50       ` Reinette Chatre
2024-11-22 21:04         ` Moger, Babu
2024-11-22 22:07           ` Reinette Chatre
2024-11-23  0:09             ` Moger, Babu
2024-12-04  4:16   ` Fenghua Yu
2024-10-29 23:21 ` [PATCH v9 19/26] x86/resctrl: Add the interface to unassign a MBM counter Babu Moger
2024-11-04 14:16   ` Peter Newman
2024-11-04 18:21     ` Moger, Babu
2024-11-05 10:35       ` Peter Newman
2024-11-05 19:58         ` Moger, Babu
2024-10-29 23:21 ` [PATCH v9 20/26] x86/resctrl: Auto assign/unassign counters when mbm_cntr_assign is enabled Babu Moger
2024-11-18 17:18   ` Reinette Chatre
2024-11-22  0:22     ` Moger, Babu
2024-11-22  0:26       ` Moger, Babu
2024-11-22 18:12         ` Reinette Chatre
2024-11-22 21:34           ` Moger, Babu
2024-12-04  4:16   ` Fenghua Yu
2024-12-04 17:03     ` Reinette Chatre
2024-12-04 17:14     ` Moger, Babu
2024-12-04 17:19       ` Moger, Babu
2024-10-29 23:21 ` [PATCH v9 21/26] x86/resctrl: Report "Unassigned" for MBM events in mbm_cntr_assign mode Babu Moger
2024-11-18 17:39   ` Reinette Chatre
2024-11-20 19:14     ` Moger, Babu
2024-10-29 23:21 ` [PATCH v9 22/26] x86/resctrl: Introduce the interface to switch between monitor modes Babu Moger
2024-10-29 23:21 ` [PATCH v9 23/26] x86/resctrl: Configure mbm_cntr_assign mode if supported Babu Moger
2024-11-18 19:23   ` Reinette Chatre
2024-11-20 18:59     ` Moger, Babu
2024-10-29 23:21 ` [PATCH v9 24/26] x86/resctrl: Update assignments on event configuration changes Babu Moger
2024-11-18 19:43   ` Reinette Chatre
2024-11-21  2:14     ` Moger, Babu
2024-11-21 20:58       ` Reinette Chatre
2024-11-22 20:12         ` Moger, Babu
2024-10-29 23:21 ` [PATCH v9 25/26] x86/resctrl: Introduce interface to list assignment states of all the groups Babu Moger
2024-10-29 23:21 ` [PATCH v9 26/26] x86/resctrl: Introduce interface to modify assignment states of " Babu Moger
2024-11-18 21:51   ` Reinette Chatre
2024-11-21 20:29     ` Moger, Babu

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=0eff9462-e7e2-49a9-9538-c8907024322f@amd.com \
    --to=bmoger@amd.com \
    --cc=akpm@linux-foundation.org \
    --cc=andrew.cooper3@citrix.com \
    --cc=babu.moger@amd.com \
    --cc=bp@alien8.de \
    --cc=brijesh.singh@amd.com \
    --cc=corbet@lwn.net \
    --cc=daniel.sneddon@linux.intel.com \
    --cc=dave.hansen@linux.intel.com \
    --cc=ebiggers@google.com \
    --cc=eranian@google.com \
    --cc=fenghua.yu@intel.com \
    --cc=hpa@zytor.com \
    --cc=james.morse@arm.com \
    --cc=jithu.joseph@intel.com \
    --cc=jpoimboe@kernel.org \
    --cc=kai.huang@intel.com \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=maciej.wieczor-retman@intel.com \
    --cc=mario.limonciello@amd.com \
    --cc=mingo@redhat.com \
    --cc=paulmck@kernel.org \
    --cc=pawan.kumar.gupta@linux.intel.com \
    --cc=perry.yuan@amd.com \
    --cc=peternewman@google.com \
    --cc=reinette.chatre@intel.com \
    --cc=rostedt@goodmis.org \
    --cc=sandipan.das@amd.com \
    --cc=seanjc@google.com \
    --cc=tan.shaopeng@fujitsu.com \
    --cc=tglx@linutronix.de \
    --cc=thomas.lendacky@amd.com \
    --cc=thuth@redhat.com \
    --cc=tony.luck@intel.com \
    --cc=x86@kernel.org \
    --cc=xiaoyao.li@intel.com \
    --cc=xin3.li@intel.com \
    --cc=xiongwei.song@windriver.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).