patches.lists.linux.dev archive mirror
 help / color / mirror / Atom feed
From: Tony Luck <tony.luck@intel.com>
To: Fenghua Yu <fenghuay@nvidia.com>,
	Reinette Chatre <reinette.chatre@intel.com>,
	Maciej Wieczor-Retman <maciej.wieczor-retman@intel.com>,
	Peter Newman <peternewman@google.com>,
	James Morse <james.morse@arm.com>,
	Babu Moger <babu.moger@amd.com>,
	Drew Fustini <dfustini@baylibre.com>,
	Dave Martin <Dave.Martin@arm.com>,
	Anil Keshavamurthy <anil.s.keshavamurthy@intel.com>,
	Chen Yu <yu.c.chen@intel.com>
Cc: x86@kernel.org, linux-kernel@vger.kernel.org,
	patches@lists.linux.dev, Tony Luck <tony.luck@intel.com>
Subject: [PATCH v4 04/31] fs/resctrl: Change how and when events are initialized
Date: Mon, 28 Apr 2025 17:33:30 -0700	[thread overview]
Message-ID: <20250429003359.375508-5-tony.luck@intel.com> (raw)
In-Reply-To: <20250429003359.375508-1-tony.luck@intel.com>

Existing code assumes that all monitor events are associated with
the RDT_RESOURCE_L3 resource. Also that all event enumeration is
complete during early resctrl initialization. Neither of these
assumptions remain true for new events.

Each resource must include a list of enabled events that is used
to add appropriately named files when creating mon_data directories
and to for the contents of "info/{resource}_MON/mon_features" file.

Move the building of enabled event lists for each resource from
resctrl_mon_resource_init() to rdt_get_tree() to delay it until
mount of the resctrl file system.

Add a new field to struct mon_evt to record which resource each
event is associated with so that events are added to the correct
resource event list.

Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 fs/resctrl/internal.h |  4 ++++
 fs/resctrl/monitor.c  | 33 ++++++++++++++++++++++-----------
 fs/resctrl/rdtgroup.c |  2 ++
 3 files changed, 28 insertions(+), 11 deletions(-)

diff --git a/fs/resctrl/internal.h b/fs/resctrl/internal.h
index 6029b3285dd3..b69170760316 100644
--- a/fs/resctrl/internal.h
+++ b/fs/resctrl/internal.h
@@ -68,6 +68,7 @@ static inline struct rdt_fs_context *rdt_fc2context(struct fs_context *fc)
 /**
  * struct mon_evt - Entry in the event list of a resource
  * @evtid:		event id
+ * @rid:		index of the resource for this event
  * @name:		name of the event
  * @configurable:	true if the event is configurable
  * @enabled:		true if the event is enabled
@@ -75,6 +76,7 @@ static inline struct rdt_fs_context *rdt_fc2context(struct fs_context *fc)
  */
 struct mon_evt {
 	enum resctrl_event_id	evtid;
+	enum resctrl_res_level	rid;
 	char			*name;
 	bool			configurable;
 	bool			enabled;
@@ -397,6 +399,8 @@ enum resctrl_event_id resctrl_get_mon_event_by_name(char *name);
 
 char *resctrl_mon_event_name(enum resctrl_event_id evt);
 
+void resctrl_init_mon_events(void);
+
 #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 625cd328c790..a5a523f73249 100644
--- a/fs/resctrl/monitor.c
+++ b/fs/resctrl/monitor.c
@@ -845,14 +845,17 @@ struct mon_evt mon_event_all[QOS_NUM_EVENTS] = {
 	[QOS_L3_OCCUP_EVENT_ID] = {
 		.name	= "llc_occupancy",
 		.evtid	= QOS_L3_OCCUP_EVENT_ID,
+		.rid	= RDT_RESOURCE_L3,
 	},
 	[QOS_L3_MBM_TOTAL_EVENT_ID] = {
 		.name	= "mbm_total_bytes",
 		.evtid	= QOS_L3_MBM_TOTAL_EVENT_ID,
+		.rid	= RDT_RESOURCE_L3,
 	},
 	[QOS_L3_MBM_LOCAL_EVENT_ID] = {
 		.name	= "mbm_local_bytes",
 		.evtid	= QOS_L3_MBM_LOCAL_EVENT_ID,
+		.rid	= RDT_RESOURCE_L3,
 	},
 };
 
@@ -886,21 +889,31 @@ char *resctrl_mon_event_name(enum resctrl_event_id evt)
 }
 
 /*
- * Initialize the event list for the resource.
+ * Initialize the event list for all mon_capable resources.
  *
- * Note that MBM events are also part of RDT_RESOURCE_L3 resource
- * because as per the SDM the total and local memory bandwidth
- * are enumerated as part of L3 monitoring.
+ * Called on each mount of the resctrl file system when all
+ * events have been enumerated. Only needs to build the per-resource
+ * event lists once.
  */
-static void l3_mon_evt_init(struct rdt_resource *r)
+void resctrl_init_mon_events(void)
 {
 	enum resctrl_event_id evt;
+	struct rdt_resource *r;
+	static bool only_once;
+
+	if (only_once)
+		return;
+	only_once = true;
 
-	INIT_LIST_HEAD(&r->evt_list);
+	for_each_mon_capable_rdt_resource(r)
+		INIT_LIST_HEAD(&r->evt_list);
 
-	for (evt = 0; evt < QOS_NUM_EVENTS; evt++)
-		if (mon_event_all[evt].enabled)
-			list_add_tail(&mon_event_all[evt].list, &r->evt_list);
+	for (evt = 0; evt < QOS_NUM_EVENTS; evt++) {
+		if (!mon_event_all[evt].enabled)
+			continue;
+		r = resctrl_arch_get_resource(mon_event_all[evt].rid);
+		list_add_tail(&mon_event_all[evt].list, &r->evt_list);
+	}
 }
 
 /**
@@ -927,8 +940,6 @@ int resctrl_mon_resource_init(void)
 	if (ret)
 		return ret;
 
-	l3_mon_evt_init(r);
-
 	if (resctrl_arch_is_evt_configurable(QOS_L3_MBM_TOTAL_EVENT_ID)) {
 		mon_event_all[QOS_L3_MBM_TOTAL_EVENT_ID].configurable = true;
 		resctrl_file_fflags_init("mbm_total_bytes_config",
diff --git a/fs/resctrl/rdtgroup.c b/fs/resctrl/rdtgroup.c
index c06752dfcb7c..e66dc041be5f 100644
--- a/fs/resctrl/rdtgroup.c
+++ b/fs/resctrl/rdtgroup.c
@@ -2591,6 +2591,8 @@ static int rdt_get_tree(struct fs_context *fc)
 		goto out;
 	}
 
+	resctrl_init_mon_events();
+
 	ret = rdtgroup_setup_root(ctx);
 	if (ret)
 		goto out;
-- 
2.48.1


  parent reply	other threads:[~2025-04-29  0:34 UTC|newest]

Thread overview: 72+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-04-29  0:33 [PATCH v4 00/31] x86/resctrl telemetry monitoring Tony Luck
2025-04-29  0:33 ` [PATCH v4 01/31] x86,fs/resctrl: Drop rdt_mon_features variable Tony Luck
2025-05-08  3:28   ` Reinette Chatre
2025-05-08 18:32     ` Luck, Tony
2025-05-08 23:44       ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 02/31] x86,fs/resctrl: Prepare for more monitor events Tony Luck
2025-05-08  3:30   ` Reinette Chatre
2025-05-09 15:02   ` Peter Newman
2025-04-29  0:33 ` [PATCH v4 03/31] fs/resctrl: Clean up rdtgroup_mba_mbps_event_{show,write}() Tony Luck
2025-05-08  3:31   ` Reinette Chatre
2025-04-29  0:33 ` Tony Luck [this message]
2025-05-08  3:31   ` [PATCH v4 04/31] fs/resctrl: Change how and when events are initialized Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 05/31] fs/resctrl: Set up Kconfig options for telemetry events Tony Luck
2025-05-08  3:32   ` Reinette Chatre
2025-05-10  9:58   ` Chen, Yu C
2025-05-12 14:19     ` Luck, Tony
2025-04-29  0:33 ` [PATCH v4 06/31] x86/rectrl: Fake OOBMSM interface Tony Luck
2025-04-30 23:02   ` Luck, Tony
2025-05-08  3:33   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 07/31] x86,fs/resctrl: Improve domain type checking Tony Luck
2025-05-08  3:36   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 08/31] x86/resctrl: Move L3 initialization out of domain_add_cpu_mon() Tony Luck
2025-05-08  3:37   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 09/31] x86,fs/resctrl: Refactor domain_remove_cpu_mon() ready for new domain types Tony Luck
2025-05-08  3:37   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 10/31] x86/resctrl: Change generic monitor functions to use struct rdt_domain_hdr Tony Luck
2025-05-08  3:38   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 11/31] x86,fs/resctrl: Rename struct rdt_mon_domain and rdt_hw_mon_domain Tony Luck
2025-05-08  3:39   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 12/31] fs/resctrl: Improve handling for events that can be read from any CPU Tony Luck
2025-05-08  3:54   ` Reinette Chatre
2025-05-13  3:19   ` Chen, Yu C
2025-05-13 16:20     ` Luck, Tony
2025-05-14  9:11       ` Chen, Yu C
2025-04-29  0:33 ` [PATCH v4 13/31] fs/resctrl: Add support for additional monitor event display formats Tony Luck
2025-05-08 15:49   ` Reinette Chatre
2025-05-08 20:28     ` Luck, Tony
2025-05-08 23:45       ` Reinette Chatre
2025-05-09 11:29         ` Dave Martin
2025-05-09 14:46           ` Peter Newman
2025-05-09 16:38             ` Luck, Tony
2025-05-09 16:43             ` Dave Martin
2025-04-29  0:33 ` [PATCH v4 14/31] fs/resctrl: Add an architectural hook called for each mount Tony Luck
2025-05-08 15:50   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 15/31] x86/resctrl: Add and initialize rdt_resource for package scope core monitor Tony Luck
2025-05-08 15:50   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 16/31] x86/resctrl: Add first part of telemetry event enumeration Tony Luck
2025-05-08 15:53   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 17/31] x86/resctrl: Add second " Tony Luck
2025-05-08 15:54   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 18/31] x86/resctrl: Add third " Tony Luck
2025-05-08 15:56   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 19/31] x86,fs/resctrl: Fill in details of Clearwater Forest events Tony Luck
2025-05-08 15:54   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 20/31] x86/resctrl: Check for adequate MMIO space Tony Luck
2025-05-08 15:56   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 21/31] x86/resctrl: Add fourth part of telemetry event enumeration Tony Luck
2025-05-08 15:56   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 22/31] x86/resctrl: Read core telemetry events Tony Luck
2025-05-08 15:57   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 23/31] x86,fs/resctrl: Handle domain creation/deletion for RDT_RESOURCE_PERF_PKG Tony Luck
2025-05-08 15:58   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 24/31] fs/resctrl: Add type define for PERF_PKG files Tony Luck
2025-04-29  0:33 ` [PATCH v4 25/31] x86/resctrl: Final steps to enable RDT_RESOURCE_PERF_PKG Tony Luck
2025-04-29  0:33 ` [PATCH v4 26/31] x86/resctrl: Add energy/perf choices to rdt boot option Tony Luck
2025-05-08 15:58   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 27/31] x86/resctrl: Handle number of RMIDs supported by telemetry resources Tony Luck
2025-05-08 15:59   ` Reinette Chatre
2025-04-29  0:33 ` [PATCH v4 28/31] x86,fs/resctrl: Fix RMID allocation for multiple monitor resources Tony Luck
2025-04-29  0:33 ` [PATCH v4 29/31] fs/resctrl: Add interface for per-resource debug info files Tony Luck
2025-04-29  0:33 ` [PATCH v4 30/31] x86/resctrl: Add info/PERF_PKG_MON/status file Tony Luck
2025-04-29  0:33 ` [PATCH v4 31/31] x86/resctrl: Update Documentation for package events Tony Luck

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=20250429003359.375508-5-tony.luck@intel.com \
    --to=tony.luck@intel.com \
    --cc=Dave.Martin@arm.com \
    --cc=anil.s.keshavamurthy@intel.com \
    --cc=babu.moger@amd.com \
    --cc=dfustini@baylibre.com \
    --cc=fenghuay@nvidia.com \
    --cc=james.morse@arm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=maciej.wieczor-retman@intel.com \
    --cc=patches@lists.linux.dev \
    --cc=peternewman@google.com \
    --cc=reinette.chatre@intel.com \
    --cc=x86@kernel.org \
    --cc=yu.c.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;
as well as URLs for NNTP newsgroup(s).