linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Reinette Chatre <reinette.chatre@intel.com>
To: Babu Moger <babu.moger@amd.com>, <corbet@lwn.net>,
	<tony.luck@intel.com>, <james.morse@arm.com>,
	<tglx@linutronix.de>, <mingo@redhat.com>, <bp@alien8.de>,
	<dave.hansen@linux.intel.com>
Cc: <Dave.Martin@arm.com>, <x86@kernel.org>, <hpa@zytor.com>,
	<akpm@linux-foundation.org>, <paulmck@kernel.org>,
	<rostedt@goodmis.org>, <Neeraj.Upadhyay@amd.com>,
	<david@redhat.com>, <arnd@arndb.de>, <fvdl@google.com>,
	<seanjc@google.com>, <jpoimboe@kernel.org>,
	<pawan.kumar.gupta@linux.intel.com>, <xin@zytor.com>,
	<manali.shukla@amd.com>, <tao1.su@linux.intel.com>,
	<sohil.mehta@intel.com>, <kai.huang@intel.com>,
	<xiaoyao.li@intel.com>, <peterz@infradead.org>,
	<xin3.li@intel.com>, <kan.liang@linux.intel.com>,
	<mario.limonciello@amd.com>, <thomas.lendacky@amd.com>,
	<perry.yuan@amd.com>, <gautham.shenoy@amd.com>,
	<chang.seok.bae@intel.com>, <linux-doc@vger.kernel.org>,
	<linux-kernel@vger.kernel.org>, <peternewman@google.com>,
	<eranian@google.com>
Subject: Re: [PATCH v15 17/34] fs/resctrl: Add the functionality to assign MBM events
Date: Thu, 17 Jul 2025 20:47:36 -0700	[thread overview]
Message-ID: <f45d9b7c-90cc-439e-b51f-5e7aa3d8ceea@intel.com> (raw)
In-Reply-To: <70ef9637d73825ee5e7409ab04df1039dec0873c.1752013061.git.babu.moger@amd.com>

Hi Babu,

On 7/8/25 3:17 PM, Babu Moger wrote:
> When supported, "mbm_event" counter assignment mode offers "num_mbm_cntrs"
> number of counters that can be assigned to RMID, event pairs and monitor
> bandwidth usage as long as it is assigned.
> 
> Add the functionality to allocate and assign a counter to an RMID, event
> pair in the domain.
> 
> If all the counters are in use, kernel will log the error message "Unable
> to allocate counter in domain" in /sys/fs/resctrl/info/last_cmd_status
> when a new assignment is requested. Exit on the first failure when
> assigning counters across all the domains.
> 
> Signed-off-by: Babu Moger <babu.moger@amd.com>
> ---

...

> ---
>  fs/resctrl/internal.h |   3 +
>  fs/resctrl/monitor.c  | 131 ++++++++++++++++++++++++++++++++++++++++++
>  2 files changed, 134 insertions(+)
> 
> diff --git a/fs/resctrl/internal.h b/fs/resctrl/internal.h
> index ea5c9fa932aa..8879e127a8b8 100644
> --- a/fs/resctrl/internal.h
> +++ b/fs/resctrl/internal.h
> @@ -387,6 +387,9 @@ bool closid_allocated(unsigned int closid);
>  
>  int resctrl_find_cleanest_closid(void);
>  
> +int rdtgroup_assign_cntr_event(struct rdt_mon_domain *d, struct rdtgroup *rdtgrp,
> +			       struct mon_evt *mevt);
> +
>  #ifdef CONFIG_RESCTRL_FS_PSEUDO_LOCK
>  int rdtgroup_locksetup_enter(struct rdtgroup *rdtgrp);
>  
> diff --git a/fs/resctrl/monitor.c b/fs/resctrl/monitor.c
> index 11327bd8cf72..bb074773420d 100644
> --- a/fs/resctrl/monitor.c
> +++ b/fs/resctrl/monitor.c
> @@ -952,3 +952,134 @@ void resctrl_mon_resource_exit(void)
>  
>  	dom_data_exit(r);
>  }
> +
> +/*
> + * resctrl_config_cntr() - Configure the counter ID for the event, RMID pair in
> + * the domain.
> + *
> + * Assign the counter if @assign is true else unassign the counter. Reset the
> + * associated non-architectural state.

Is this API only for assignment? It seems so. Looks like resctrl_config_cntr() is used
for assign/unassign while rdtgroup_update_cntr_event()/resctrl_update_cntr_allrdtgrp() is
for re-configure.

I think this will be easier to understand if the function names and comments match this
usage. To help make clear when a counter is assigned/unassigned or when an assigned
counter is re-configured.

resctrl_config_cntr() can be renamed to rdtgroup_assign_cntr() with its description
matching what it does: (a) it does not reconfigure a counter but assign/unassign it,
and (b) it operates on the rdtgroup.
It is only the underlying arch API that uses "configure" for assign, unassign, as
well as configure.


> + */
> +static void resctrl_config_cntr(struct rdt_resource *r, struct rdt_mon_domain *d,
> +				enum resctrl_event_id evtid, u32 rmid, u32 closid,
> +				u32 cntr_id, bool assign)
> +{
> +	struct mbm_state *m;
> +
> +	resctrl_arch_config_cntr(r, d, evtid, rmid, closid, cntr_id, assign);
> +
> +	m = get_mbm_state(d, closid, rmid, evtid);
> +	if (m)
> +		memset(m, 0, sizeof(*m));
> +}
> +
> +/*
> + * mbm_cntr_get() - Return the counter ID for the matching @evtid and @rdtgrp.
> + *
> + * Return:
> + * Valid counter ID on success, or -ENOENT on failure.
> + */
> +static int mbm_cntr_get(struct rdt_resource *r, struct rdt_mon_domain *d,
> +			struct rdtgroup *rdtgrp, enum resctrl_event_id evtid)
> +{
> +	int cntr_id;
> +
> +	if (!r->mon.mbm_cntr_assignable)
> +		return -ENOENT;
> +
> +	if (!resctrl_is_mbm_event(evtid))
> +		return -ENOENT;
> +
> +	for (cntr_id = 0; cntr_id < r->mon.num_mbm_cntrs; cntr_id++) {
> +		if (d->cntr_cfg[cntr_id].rdtgrp == rdtgrp &&
> +		    d->cntr_cfg[cntr_id].evtid == evtid)
> +			return cntr_id;
> +	}
> +
> +	return -ENOENT;
> +}
> +
> +/*
> + * mbm_cntr_alloc() - Initialize and return a new counter ID in the domain @d.
> + * Caller must ensure that the specified event is not assigned already.
> + *
> + * Return:
> + * Valid counter ID on success, or -ENOSPC on failure.
> + */
> +static int mbm_cntr_alloc(struct rdt_resource *r, struct rdt_mon_domain *d,
> +			  struct rdtgroup *rdtgrp, enum resctrl_event_id evtid)
> +{
> +	int cntr_id;
> +
> +	for (cntr_id = 0; cntr_id < r->mon.num_mbm_cntrs; cntr_id++) {
> +		if (!d->cntr_cfg[cntr_id].rdtgrp) {
> +			d->cntr_cfg[cntr_id].rdtgrp = rdtgrp;
> +			d->cntr_cfg[cntr_id].evtid = evtid;
> +			return cntr_id;
> +		}
> +	}
> +
> +	return -ENOSPC;
> +}
> +
> +/*
> + * rdtgroup_alloc_config_cntr() - Allocate a counter ID and configure it for the
> + * event pointed to by @mevt and the resctrl group @rdtgrp within the domain @d.
> + *
> + * Return:
> + * 0 on success, < 0 on failure.
> + */
> +static int rdtgroup_alloc_config_cntr(struct rdt_resource *r, struct rdt_mon_domain *d,
> +				      struct rdtgroup *rdtgrp, enum resctrl_event_id evtid)

How about rdtgroup_alloc_config_cntr() -> rdtgroup_alloc_assign_cntr()?

> +{
> +	int cntr_id;
> +
> +	/* No action required if the counter is assigned already. */
> +	cntr_id = mbm_cntr_get(r, d, rdtgrp, evtid);
> +	if (cntr_id >= 0)
> +		return 0;
> +
> +	cntr_id = mbm_cntr_alloc(r, d, rdtgrp, evtid);
> +	if (cntr_id <  0) {
> +		rdt_last_cmd_printf("Unable to allocate counter in domain %d\n",
> +				    d->hdr.id);
> +		return cntr_id;
> +	}
> +
> +	resctrl_config_cntr(r, d, evtid, rdtgrp->mon.rmid, rdtgrp->closid,
> +			    cntr_id, true);
> +
> +	return 0;
> +}
> +
> +/*
> + * rdtgroup_assign_cntr_event() - Assign a hardware counter for the event in
> + * @mevt to the resctrl group @rdtgrp. Assign counters to all domains if @d is
> + * NULL; otherwise, assign the counter to the specified domain @d.
> + *
> + * If all counters in a domain are already in use, resctrl_alloc_config_cntr()

resctrl_alloc_config_cntr() needs update to match function name

> + * will fail. The assignment process will abort at the first failure encountered
> + * during domain traversal, which may result in the event being only partially
> + * assigned.
> + *
> + * Return:
> + * 0 on success, < 0 on failure.
> + */
> +int rdtgroup_assign_cntr_event(struct rdt_mon_domain *d, struct rdtgroup *rdtgrp,
> +			       struct mon_evt *mevt)
> +{
> +	struct rdt_resource *r = resctrl_arch_get_resource(mevt->rid);
> +	int ret = 0;
> +
> +	if (!d) {
> +		list_for_each_entry(d, &r->mon_domains, hdr.list) {
> +			ret = rdtgroup_alloc_config_cntr(r, d, rdtgrp, mevt->evtid);
> +			if (ret)
> +				return ret;
> +		}
> +	} else {
> +		ret = rdtgroup_alloc_config_cntr(r, d, rdtgrp, mevt->evtid);
> +	}
> +
> +	return ret;
> +}

Reinette


  reply	other threads:[~2025-07-18  3:47 UTC|newest]

Thread overview: 98+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-07-08 22:17 [PATCH v15 00/34] fs,x86/resctrl: Support AMD Assignable Bandwidth Monitoring Counters (ABMC) Babu Moger
2025-07-08 22:17 ` [PATCH v15 01/34] x86,fs/resctrl: Consolidate monitor event descriptions Babu Moger
2025-07-17 18:43   ` Reinette Chatre
2025-07-18 14:19     ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 02/34] x86,fs/resctrl: Replace architecture event enabled checks Babu Moger
2025-07-08 22:17 ` [PATCH v15 03/34] x86/resctrl: Remove 'rdt_mon_features' global variable Babu Moger
2025-07-08 22:17 ` [PATCH v15 04/34] x86,fs/resctrl: Prepare for more monitor events Babu Moger
2025-07-08 22:17 ` [PATCH v15 05/34] x86/cpufeatures: Add support for Assignable Bandwidth Monitoring Counters (ABMC) Babu Moger
2025-07-15 11:47   ` Peter Newman
2025-07-15 14:16     ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 06/34] x86/resctrl: Add ABMC feature in the command line options Babu Moger
2025-07-17 18:44   ` Reinette Chatre
2025-07-08 22:17 ` [PATCH v15 07/34] x86,fs/resctrl: Consolidate monitoring related data from rdt_resource Babu Moger
2025-07-17 18:44   ` Reinette Chatre
2025-07-08 22:17 ` [PATCH v15 08/34] x86,fs/resctrl: Detect Assignable Bandwidth Monitoring feature details Babu Moger
2025-07-15 16:26   ` Reinette Chatre
2025-07-15 16:53     ` Moger, Babu
2025-07-17 18:45   ` Reinette Chatre
2025-07-21 15:20     ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 09/34] x86/resctrl: Add support to enable/disable AMD ABMC feature Babu Moger
2025-07-17 18:46   ` Reinette Chatre
2025-07-08 22:17 ` [PATCH v15 10/34] fs/resctrl: Introduce the interface to display monitoring modes Babu Moger
2025-07-17 18:46   ` Reinette Chatre
2025-07-08 22:17 ` [PATCH v15 11/34] fs/resctrl: Add resctrl file to display number of assignable counters Babu Moger
2025-07-17 18:46   ` Reinette Chatre
2025-07-08 22:17 ` [PATCH v15 12/34] fs/resctrl: Introduce mbm_cntr_cfg to track assignable counters per domain Babu Moger
2025-07-17 18:46   ` Reinette Chatre
2025-07-08 22:17 ` [PATCH v15 13/34] fs/resctrl: Introduce interface to display number of free MBM counters Babu Moger
2025-07-17 18:47   ` Reinette Chatre
2025-07-08 22:17 ` [PATCH v15 14/34] x86/resctrl: Add data structures and definitions for ABMC assignment Babu Moger
2025-07-17 18:47   ` Reinette Chatre
2025-07-08 22:17 ` [PATCH v15 15/34] fs/resctrl: Introduce event configuration field in struct mon_evt Babu Moger
2025-07-17 18:47   ` Reinette Chatre
2025-07-08 22:17 ` [PATCH v15 16/34] x86,fs/resctrl: Implement resctrl_arch_config_cntr() to assign a counter with ABMC Babu Moger
2025-07-17 18:49   ` Reinette Chatre
2025-07-21 17:40     ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 17/34] fs/resctrl: Add the functionality to assign MBM events Babu Moger
2025-07-18  3:47   ` Reinette Chatre [this message]
2025-07-21 19:54     ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 18/34] fs/resctrl: Add the functionality to unassign " Babu Moger
2025-07-18  3:48   ` Reinette Chatre
2025-07-21 20:21     ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 19/34] fs/resctrl: Pass struct rdtgroup instead of individual members Babu Moger
2025-07-18  3:54   ` Reinette Chatre
2025-07-21 20:59     ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 20/34] fs/resctrl: Introduce counter read, reset calls in mbm_event mode Babu Moger
2025-07-18  3:50   ` Reinette Chatre
2025-07-21 23:39     ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 21/34] x86/resctrl: Refactor resctrl_arch_rmid_read() Babu Moger
2025-07-18  3:51   ` Reinette Chatre
2025-07-22 14:23     ` Moger, Babu
2025-07-22 14:56       ` Reinette Chatre
2025-07-22 15:25         ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 22/34] x86/resctrl: Implement resctrl_arch_reset_cntr() and resctrl_arch_cntr_read() Babu Moger
2025-07-18  3:51   ` Reinette Chatre
2025-07-22 15:51     ` Moger, Babu
2025-07-22 23:27       ` Reinette Chatre
2025-07-23 16:48         ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 23/34] fs/resctrl: Support counter read/reset with mbm_event assignment mode Babu Moger
2025-07-18  3:53   ` Reinette Chatre
2025-07-22 17:53     ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 24/34] fs/resctrl: Report 'Unassigned' for MBM events in mbm_event mode Babu Moger
2025-07-18  3:53   ` Reinette Chatre
2025-07-22 18:15     ` Moger, Babu
2025-07-22 23:28       ` Reinette Chatre
2025-07-23  0:26         ` Moger, Babu
2025-07-23  2:05           ` Reinette Chatre
2025-07-23 13:14             ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 25/34] fs/resctrl: Add definitions for MBM event configuration Babu Moger
2025-07-18  3:55   ` Reinette Chatre
2025-07-22 19:34     ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 26/34] fs/resctrl: Add event configuration directory under info/L3_MON/ Babu Moger
2025-07-18  3:54   ` Reinette Chatre
2025-07-18 22:20     ` Reinette Chatre
2025-07-22 20:22       ` Moger, Babu
2025-07-22 20:11     ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 27/34] fs/resctrl: Provide interface to update the event configurations Babu Moger
2025-07-18  3:55   ` Reinette Chatre
2025-07-22 22:55     ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 28/34] fs/resctrl: Introduce mbm_assign_on_mkdir to enable assignments on mkdir Babu Moger
2025-07-15 13:53   ` Peter Newman
2025-07-15 14:18     ` Moger, Babu
2025-07-15 14:27       ` Reinette Chatre
2025-07-15 15:28         ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 29/34] x86,fs/resctrl: Auto assign counters on mkdir and clean up on group removal Babu Moger
2025-07-18  3:56   ` Reinette Chatre
2025-07-22 23:59     ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 30/34] fs/resctrl: Introduce mbm_L3_assignments to list assignments in a group Babu Moger
2025-07-08 22:17 ` [PATCH v15 31/34] fs/resctrl: Introduce the interface to modify " Babu Moger
2025-07-18  4:01   ` Reinette Chatre
2025-07-23 16:19     ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 32/34] fs/resctrl: Disable BMEC event configuration when mbm_event mode is enabled Babu Moger
2025-07-18  4:02   ` Reinette Chatre
2025-07-23 17:30     ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 33/34] fs/resctrl: Introduce the interface to switch between monitor modes Babu Moger
2025-07-18  4:03   ` Reinette Chatre
2025-07-23 18:50     ` Moger, Babu
2025-07-08 22:17 ` [PATCH v15 34/34] x86/resctrl: Configure mbm_event mode if supported Babu Moger

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=f45d9b7c-90cc-439e-b51f-5e7aa3d8ceea@intel.com \
    --to=reinette.chatre@intel.com \
    --cc=Dave.Martin@arm.com \
    --cc=Neeraj.Upadhyay@amd.com \
    --cc=akpm@linux-foundation.org \
    --cc=arnd@arndb.de \
    --cc=babu.moger@amd.com \
    --cc=bp@alien8.de \
    --cc=chang.seok.bae@intel.com \
    --cc=corbet@lwn.net \
    --cc=dave.hansen@linux.intel.com \
    --cc=david@redhat.com \
    --cc=eranian@google.com \
    --cc=fvdl@google.com \
    --cc=gautham.shenoy@amd.com \
    --cc=hpa@zytor.com \
    --cc=james.morse@arm.com \
    --cc=jpoimboe@kernel.org \
    --cc=kai.huang@intel.com \
    --cc=kan.liang@linux.intel.com \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=manali.shukla@amd.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=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=seanjc@google.com \
    --cc=sohil.mehta@intel.com \
    --cc=tao1.su@linux.intel.com \
    --cc=tglx@linutronix.de \
    --cc=thomas.lendacky@amd.com \
    --cc=tony.luck@intel.com \
    --cc=x86@kernel.org \
    --cc=xiaoyao.li@intel.com \
    --cc=xin3.li@intel.com \
    --cc=xin@zytor.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).