LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Matt Fleming <matt@codeblueprint.co.uk>
To: Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@kernel.org>, Jiri Olsa <jolsa@redhat.com>,
	Arnaldo Carvalho de Melo <acme@kernel.org>,
	Andi Kleen <andi@firstfloor.org>
Cc: Thomas Gleixner <tglx@linutronix.de>,
	linux-kernel@vger.kernel.org, "H. Peter Anvin" <hpa@zytor.com>,
	Kanaka Juvva <kanaka.d.juvva@intel.com>,
	Vikas Shivappa <vikas.shivappa@linux.intel.com>,
	Matt Fleming <matt.fleming@intel.com>
Subject: [PATCH v5 9/9] perf/x86/intel: Enable conflicting event scheduling for CQM
Date: Fri, 23 Jan 2015 18:45:48 +0000	[thread overview]
Message-ID: <1422038748-21397-10-git-send-email-matt@codeblueprint.co.uk> (raw)
In-Reply-To: <1422038748-21397-1-git-send-email-matt@codeblueprint.co.uk>

From: Matt Fleming <matt.fleming@intel.com>

We can leverage the workqueue that we use for RMID rotation to support
scheduling of conflicting monitoring events. Allowing events that
monitor conflicting things is done at various other places in the perf
subsystem, so there's precedent there.

An example of two conflicting events would be monitoring a cgroup and
simultaneously monitoring a task within that cgroup.

This uses the cache_groups list as a queuing mechanism, where every
event that reaches the front of the list gets the chance to be scheduled
in, possibly descheduling any conflicting events that are running.

Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
---
Changes in v5:

 - s/intel_cqm_sched_out_events/intel_sched_out_conflicting_events/

 - Drop list_move_tail() which breaks rotatation fairness

 arch/x86/kernel/cpu/perf_event_intel_cqm.c | 130 +++++++++++++++++++----------
 1 file changed, 84 insertions(+), 46 deletions(-)

diff --git a/arch/x86/kernel/cpu/perf_event_intel_cqm.c b/arch/x86/kernel/cpu/perf_event_intel_cqm.c
index 3b3e84985b9e..e6c6c4834629 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_cqm.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_cqm.c
@@ -507,7 +507,6 @@ static unsigned int __rmid_queue_time_ms = RMID_DEFAULT_QUEUE_TIME;
 static bool intel_cqm_rmid_stabilize(unsigned int *available)
 {
 	struct cqm_rmid_entry *entry, *tmp;
-	struct perf_event *event;
 
 	lockdep_assert_held(&cache_mutex);
 
@@ -577,19 +576,9 @@ static bool intel_cqm_rmid_stabilize(unsigned int *available)
 
 		/*
 		 * If we have groups waiting for RMIDs, hand
-		 * them one now.
+		 * them one now provided they don't conflict.
 		 */
-		list_for_each_entry(event, &cache_groups,
-				    hw.cqm_groups_entry) {
-			if (__rmid_valid(event->hw.cqm_rmid))
-				continue;
-
-			intel_cqm_xchg_rmid(event, entry->rmid);
-			entry = NULL;
-			break;
-		}
-
-		if (!entry)
+		if (intel_cqm_sched_in_event(entry->rmid))
 			continue;
 
 		/*
@@ -604,25 +593,73 @@ static bool intel_cqm_rmid_stabilize(unsigned int *available)
 
 /*
  * Pick a victim group and move it to the tail of the group list.
+ * @next: The first group without an RMID
  */
-static struct perf_event *
-__intel_cqm_pick_and_rotate(void)
+static void __intel_cqm_pick_and_rotate(struct perf_event *next)
 {
 	struct perf_event *rotor;
+	unsigned int rmid;
 
 	lockdep_assert_held(&cache_mutex);
-	lockdep_assert_held(&cache_lock);
 
 	rotor = list_first_entry(&cache_groups, struct perf_event,
 				 hw.cqm_groups_entry);
+
+	/*
+	 * The group at the front of the list should always have a valid
+	 * RMID. If it doesn't then no groups have RMIDs assigned and we
+	 * don't need to rotate the list.
+	 */
+	if (next == rotor)
+		return;
+
+	rmid = intel_cqm_xchg_rmid(rotor, INVALID_RMID);
+	__put_rmid(rmid);
+
 	list_rotate_left(&cache_groups);
+}
+
+/*
+ * Deallocate the RMIDs from any events that conflict with @event, and
+ * place them on the back of the group list.
+ */
+static void intel_cqm_sched_out_conflicting_events(struct perf_event *event)
+{
+	struct perf_event *group, *g;
+	unsigned int rmid;
+
+	lockdep_assert_held(&cache_mutex);
+
+	list_for_each_entry_safe(group, g, &cache_groups, hw.cqm_groups_entry) {
+		if (group == event)
+			continue;
+
+		rmid = group->hw.cqm_rmid;
+
+		/*
+		 * Skip events that don't have a valid RMID.
+		 */
+		if (!__rmid_valid(rmid))
+			continue;
+
+		/*
+		 * No conflict? No problem! Leave the event alone.
+		 */
+		if (!__conflict_event(group, event))
+			continue;
 
-	return rotor;
+		intel_cqm_xchg_rmid(group, INVALID_RMID);
+		__put_rmid(rmid);
+	}
 }
 
 /*
  * Attempt to rotate the groups and assign new RMIDs.
  *
+ * We rotate for two reasons,
+ *   1. To handle the scheduling of conflicting events
+ *   2. To recycle RMIDs
+ *
  * Rotating RMIDs is complicated because the hardware doesn't give us
  * any clues.
  *
@@ -642,11 +679,10 @@ __intel_cqm_pick_and_rotate(void)
  */
 static bool __intel_cqm_rmid_rotate(void)
 {
-	struct perf_event *group, *rotor, *start = NULL;
+	struct perf_event *group, *start = NULL;
 	unsigned int threshold_limit;
 	unsigned int nr_needed = 0;
 	unsigned int nr_available;
-	unsigned int rmid;
 	bool rotated = false;
 
 	mutex_lock(&cache_mutex);
@@ -678,7 +714,9 @@ again:
 		goto stabilize;
 
 	/*
-	 * We have more event groups without RMIDs than available RMIDs.
+	 * We have more event groups without RMIDs than available RMIDs,
+	 * or we have event groups that conflict with the ones currently
+	 * scheduled.
 	 *
 	 * We force deallocate the rmid of the group at the head of
 	 * cache_groups. The first event group without an RMID then gets
@@ -688,15 +726,7 @@ again:
 	 * Rotate the cache_groups list so the previous head is now the
 	 * tail.
 	 */
-	rotor = __intel_cqm_pick_and_rotate();
-	rmid = intel_cqm_xchg_rmid(rotor, INVALID_RMID);
-
-	/*
-	 * The group at the front of the list should always have a valid
-	 * RMID. If it doesn't then no groups have RMIDs assigned.
-	 */
-	if (!__rmid_valid(rmid))
-		goto stabilize;
+	__intel_cqm_pick_and_rotate(start);
 
 	/*
 	 * If the rotation is going to succeed, reduce the threshold so
@@ -704,14 +734,14 @@ again:
 	 */
 	if (__rmid_valid(intel_cqm_rotation_rmid)) {
 		intel_cqm_xchg_rmid(start, intel_cqm_rotation_rmid);
-		intel_cqm_rotation_rmid = INVALID_RMID;
+		intel_cqm_rotation_rmid = __get_rmid();
+
+		intel_cqm_sched_out_conflicting_events(start);
 
 		if (__intel_cqm_threshold)
 			__intel_cqm_threshold--;
 	}
 
-	__put_rmid(rmid);
-
 	rotated = true;
 
 stabilize:
@@ -794,25 +824,37 @@ static void intel_cqm_rmid_rotate(struct work_struct *work)
  *
  * If we're part of a group, we use the group's RMID.
  */
-static int intel_cqm_setup_event(struct perf_event *event,
-				 struct perf_event **group)
+static void intel_cqm_setup_event(struct perf_event *event,
+				  struct perf_event **group)
 {
 	struct perf_event *iter;
+	unsigned int rmid;
+	bool conflict = false;
 
 	list_for_each_entry(iter, &cache_groups, hw.cqm_groups_entry) {
+		rmid = iter->hw.cqm_rmid;
+
 		if (__match_event(iter, event)) {
 			/* All tasks in a group share an RMID */
-			event->hw.cqm_rmid = iter->hw.cqm_rmid;
+			event->hw.cqm_rmid = rmid;
 			*group = iter;
-			return 0;
+			return;
 		}
 
-		if (__conflict_event(iter, event))
-			return -EBUSY;
+		/*
+		 * We only care about conflicts for events that are
+		 * actually scheduled in (and hence have a valid RMID).
+		 */
+		if (__conflict_event(iter, event) && __rmid_valid(rmid))
+			conflict = true;
 	}
 
-	event->hw.cqm_rmid = __get_rmid();
-	return 0;
+	if (conflict)
+		rmid = INVALID_RMID;
+	else
+		rmid = __get_rmid();
+
+	event->hw.cqm_rmid = rmid;
 }
 
 static void intel_cqm_event_read(struct perf_event *event)
@@ -1030,7 +1072,6 @@ static int intel_cqm_event_init(struct perf_event *event)
 {
 	struct perf_event *group = NULL;
 	bool rotate = false;
-	int err;
 
 	if (event->attr.type != intel_cqm_pmu.type)
 		return -ENOENT;
@@ -1056,9 +1097,7 @@ static int intel_cqm_event_init(struct perf_event *event)
 	mutex_lock(&cache_mutex);
 
 	/* Will also set rmid */
-	err = intel_cqm_setup_event(event, &group);
-	if (err)
-		goto out;
+	intel_cqm_setup_event(event, &group);
 
 	if (group) {
 		list_add_tail(&event->hw.cqm_group_entry,
@@ -1078,13 +1117,12 @@ static int intel_cqm_event_init(struct perf_event *event)
 			rotate = true;
 	}
 
-out:
 	mutex_unlock(&cache_mutex);
 
 	if (rotate)
 		schedule_delayed_work(&intel_cqm_rmid_work, 0);
 
-	return err;
+	return 0;
 }
 
 EVENT_ATTR_STR(llc_occupancy, intel_cqm_llc, "event=0x01");
-- 
1.9.3


  parent reply	other threads:[~2015-01-23 18:46 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-01-23 18:45 [PATCH v5 0/9] perf: Intel Cache QoS Monitoring support Matt Fleming
2015-01-23 18:45 ` [PATCH 1/9] perf: Make perf_cgroup_from_task() global Matt Fleming
2015-02-26  4:14   ` [tip:perf/x86] " tip-bot for Matt Fleming
2015-01-23 18:45 ` [PATCH 2/9] perf: Add ->count() function to read per-package counters Matt Fleming
2015-02-26  4:14   ` [tip:perf/x86] " tip-bot for Matt Fleming
2015-01-23 18:45 ` [PATCH 3/9] perf: Move cgroup init before PMU ->event_init() Matt Fleming
2015-02-26  4:14   ` [tip:perf/x86] " tip-bot for Matt Fleming
2015-01-23 18:45 ` [PATCH 4/9] x86: Add support for Intel Cache QoS Monitoring (CQM) detection Matt Fleming
2015-02-26  4:14   ` [tip:perf/x86] x86: Add support for Intel Cache QoS Monitoring ( CQM) detection tip-bot for Peter P Waskiewicz Jr
2015-01-23 18:45 ` [PATCH 5/9] perf/x86/intel: Add Intel Cache QoS Monitoring support Matt Fleming
2015-01-25 18:34   ` Jiri Olsa
2015-01-25 23:07     ` Matt Fleming
2015-02-26  4:15   ` [tip:perf/x86] " tip-bot for Matt Fleming
2015-02-26 18:47     ` Matt Fleming
2015-02-26 19:40       ` Peter Zijlstra
2015-03-23 12:25       ` [tip:perf/x86] perf/x86/intel: Fix Makefile to actually build the cqm driver tip-bot for Matt Fleming
2015-01-23 18:45 ` [PATCH 6/9] perf/x86/intel: Implement LRU monitoring ID allocation for CQM Matt Fleming
2015-02-26  4:15   ` [tip:perf/x86] " tip-bot for Matt Fleming
2015-01-23 18:45 ` [PATCH 7/9] perf/x86/intel: Support task events with Intel CQM Matt Fleming
2015-02-26  4:15   ` [tip:perf/x86] " tip-bot for Matt Fleming
2015-03-04  4:16     ` Vince Weaver
2015-03-05  0:55       ` Ingo Molnar
2015-03-05 21:10         ` Peter Zijlstra
2015-03-23 12:26           ` [tip:perf/x86] perf: Remove type specific target pointers tip-bot for Peter Zijlstra
2015-01-23 18:45 ` [PATCH v5 8/9] perf/x86/intel: Perform rotation on Intel CQM RMIDs Matt Fleming
2015-02-26  4:16   ` [tip:perf/x86] " tip-bot for Matt Fleming
2015-01-23 18:45 ` Matt Fleming [this message]
2015-02-26  4:16   ` [tip:perf/x86] perf/x86/intel: Enable conflicting event scheduling for CQM tip-bot for Matt Fleming

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=1422038748-21397-10-git-send-email-matt@codeblueprint.co.uk \
    --to=matt@codeblueprint.co.uk \
    --cc=acme@kernel.org \
    --cc=andi@firstfloor.org \
    --cc=hpa@zytor.com \
    --cc=jolsa@redhat.com \
    --cc=kanaka.d.juvva@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=matt.fleming@intel.com \
    --cc=mingo@kernel.org \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    --cc=vikas.shivappa@linux.intel.com \
    --subject='Re: [PATCH v5 9/9] perf/x86/intel: Enable conflicting event scheduling for CQM' \
    /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

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).