mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: mingo@redhat.com, juri.lelli@redhat.com,
	vincent.guittot@linaro.org, dietmar.eggemann@arm.com,
	rostedt@goodmis.org, bsegall@google.com, mgorman@suse.de,
	vschneid@redhat.com, clm@meta.com
Cc: linux-kernel@vger.kernel.org, peterz@infradead.org
Subject: [PATCH v2 05/12] sched: Add ttwu_queue controls
Date: Wed, 02 Jul 2025 13:49:29 +0200	[thread overview]
Message-ID: <20250702121158.817814031@infradead.org> (raw)
In-Reply-To: <20250702114924.091581796@infradead.org>

There are two (soon three) callers of ttwu_queue_wakelist(),
distinguish them with their own WF_ and add some knobs on.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20250520101727.874587738@infradead.org
---
 kernel/sched/core.c     |   22 ++++++++++++----------
 kernel/sched/features.h |    2 ++
 kernel/sched/sched.h    |    2 ++
 3 files changed, 16 insertions(+), 10 deletions(-)

--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -3888,7 +3888,7 @@ bool cpus_share_resources(int this_cpu,
 	return per_cpu(sd_share_id, this_cpu) == per_cpu(sd_share_id, that_cpu);
 }
 
-static inline bool ttwu_queue_cond(struct task_struct *p, int cpu)
+static inline bool ttwu_queue_cond(struct task_struct *p, int cpu, bool def)
 {
 	/* See SCX_OPS_ALLOW_QUEUED_WAKEUP. */
 	if (!scx_allow_ttwu_queue(p))
@@ -3929,18 +3929,19 @@ static inline bool ttwu_queue_cond(struc
 	if (!cpu_rq(cpu)->nr_running)
 		return true;
 
-	return false;
+	return def;
 }
 
 static bool ttwu_queue_wakelist(struct task_struct *p, int cpu, int wake_flags)
 {
-	if (sched_feat(TTWU_QUEUE) && ttwu_queue_cond(p, cpu)) {
-		sched_clock_cpu(cpu); /* Sync clocks across CPUs */
-		__ttwu_queue_wakelist(p, cpu, wake_flags);
-		return true;
-	}
+	bool def = sched_feat(TTWU_QUEUE_DEFAULT);
+
+	if (!ttwu_queue_cond(p, cpu, def))
+		return false;
 
-	return false;
+	sched_clock_cpu(cpu); /* Sync clocks across CPUs */
+	__ttwu_queue_wakelist(p, cpu, wake_flags);
+	return true;
 }
 
 static void ttwu_queue(struct task_struct *p, int cpu, int wake_flags)
@@ -3948,7 +3949,7 @@ static void ttwu_queue(struct task_struc
 	struct rq *rq = cpu_rq(cpu);
 	struct rq_flags rf;
 
-	if (ttwu_queue_wakelist(p, cpu, wake_flags))
+	if (sched_feat(TTWU_QUEUE) && ttwu_queue_wakelist(p, cpu, wake_flags))
 		return;
 
 	rq_lock(rq, &rf);
@@ -4251,7 +4252,8 @@ int try_to_wake_up(struct task_struct *p
 		 * scheduling.
 		 */
 		if (smp_load_acquire(&p->on_cpu) &&
-		    ttwu_queue_wakelist(p, task_cpu(p), wake_flags))
+		    sched_feat(TTWU_QUEUE_ON_CPU) &&
+		    ttwu_queue_wakelist(p, task_cpu(p), wake_flags | WF_ON_CPU))
 			break;
 
 		cpu = select_task_rq(p, p->wake_cpu, &wake_flags);
--- a/kernel/sched/features.h
+++ b/kernel/sched/features.h
@@ -81,6 +81,8 @@ SCHED_FEAT(TTWU_QUEUE, false)
  */
 SCHED_FEAT(TTWU_QUEUE, true)
 #endif
+SCHED_FEAT(TTWU_QUEUE_ON_CPU, true)
+SCHED_FEAT(TTWU_QUEUE_DEFAULT, false)
 
 /*
  * When doing wakeups, attempt to limit superfluous scans of the LLC domain.
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -2279,6 +2279,8 @@ static inline int task_on_rq_migrating(s
 #define WF_CURRENT_CPU		0x40 /* Prefer to move the wakee to the current CPU. */
 #define WF_RQ_SELECTED		0x80 /* ->select_task_rq() was called */
 
+#define WF_ON_CPU		0x0100
+
 static_assert(WF_EXEC == SD_BALANCE_EXEC);
 static_assert(WF_FORK == SD_BALANCE_FORK);
 static_assert(WF_TTWU == SD_BALANCE_WAKE);



  parent reply	other threads:[~2025-07-02 12:13 UTC|newest]

Thread overview: 101+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-07-02 11:49 [PATCH v2 00/12] sched: Address schbench regression Peter Zijlstra
2025-07-02 11:49 ` [PATCH v2 01/12] sched/psi: Optimize psi_group_change() cpu_clock() usage Peter Zijlstra
2025-07-15 19:11   ` Chris Mason
2025-07-16  6:06     ` K Prateek Nayak
2025-07-16  6:53     ` Beata Michalska
2025-07-16 10:40       ` Peter Zijlstra
2025-07-16 14:54         ` Johannes Weiner
2025-07-16 16:27         ` Chris Mason
2025-07-23  4:16         ` Aithal, Srikanth
2025-07-25  5:13         ` K Prateek Nayak
2025-07-02 11:49 ` [PATCH v2 02/12] sched/deadline: Less agressive dl_server handling Peter Zijlstra
2025-07-02 16:12   ` Juri Lelli
2025-07-10 12:46   ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2025-07-14 22:56   ` [PATCH v2 02/12] " Mel Gorman
2025-07-15 14:55     ` Chris Mason
2025-07-16 18:19       ` Mel Gorman
2025-07-30  9:34   ` Geert Uytterhoeven
2025-07-30  9:46     ` Juri Lelli
2025-07-30 10:05       ` Geert Uytterhoeven
2025-08-05 22:03   ` Chris Bainbridge
2025-08-05 23:04     ` Chris Bainbridge
2025-09-15 22:29   ` John Stultz
2025-09-16  4:18     ` John Stultz
2025-09-16  5:28       ` [RFC][PATCH] sched/deadline: Fix dl_server getting stuck, allowing cpu starvation John Stultz
2025-09-16  8:51         ` Juri Lelli
2025-09-16 11:01           ` Peter Zijlstra
2025-09-16 12:52             ` Juri Lelli
2025-09-16 14:30               ` Peter Zijlstra
2025-09-16 17:35             ` John Stultz
2025-09-16 21:30               ` Peter Zijlstra
2025-09-17  3:29                 ` John Stultz
2025-09-17  9:34                   ` Peter Zijlstra
2025-09-17 12:26                     ` Peter Zijlstra
2025-09-17 13:56                       ` Juri Lelli
2025-09-17 17:30                         ` Peter Zijlstra
2025-09-18  8:37                           ` Juri Lelli
2025-09-18  9:04                             ` Peter Zijlstra
2025-09-18  9:42                               ` Juri Lelli
2025-09-17 19:29                       ` John Stultz
2025-09-18  6:56                       ` [tip: sched/urgent] sched/deadline: Fix dl_server behaviour tip-bot2 for Peter Zijlstra
2025-09-25  7:55                       ` tip-bot2 for Peter Zijlstra
2025-09-18  6:56             ` [tip: sched/urgent] sched/deadline: Fix dl_server getting stuck tip-bot2 for Peter Zijlstra
2025-09-18 14:46               ` Dietmar Eggemann
     [not found]               ` <CGME20250922215704eucas1p1f53a65a5cd1eafd3e0db006653231efd@eucas1p1.samsung.com>
2025-09-22 21:57                 ` Marek Szyprowski
2025-09-22 23:46                   ` John Stultz
2025-09-23  6:31                     ` Marek Szyprowski
2025-09-23  7:25                   ` Peter Zijlstra
2025-09-23  7:52                     ` Marek Szyprowski
2025-09-23 22:02                   ` Peter Zijlstra
2025-09-29 15:19                     ` Marek Szyprowski
     [not found]                     ` <eae77bd0-d874-4ddf-88d7-c1ab75358f91@samsung.com>
2025-10-09  8:35                       ` Krzysztof Kozlowski
2025-10-09  9:26                       ` Peter Zijlstra
2025-10-09 11:42                         ` Marek Szyprowski
2025-09-25  7:55             ` tip-bot2 for Peter Zijlstra
2025-07-02 11:49 ` [PATCH v2 03/12] sched: Optimize ttwu() / select_task_rq() Peter Zijlstra
2025-07-10 16:47   ` Vincent Guittot
2025-07-14 22:59   ` Mel Gorman
2025-07-02 11:49 ` [PATCH v2 04/12] sched: Use lock guard in ttwu_runnable() Peter Zijlstra
2025-07-10 16:48   ` Vincent Guittot
2025-07-14 23:00   ` Mel Gorman
2025-07-02 11:49 ` Peter Zijlstra [this message]
2025-07-10 16:51   ` [PATCH v2 05/12] sched: Add ttwu_queue controls Vincent Guittot
2025-07-14 23:14   ` Mel Gorman
2025-07-02 11:49 ` [PATCH v2 06/12] sched: Introduce ttwu_do_migrate() Peter Zijlstra
2025-07-10 16:51   ` Vincent Guittot
2025-07-02 11:49 ` [PATCH v2 07/12] psi: Split psi_ttwu_dequeue() Peter Zijlstra
2025-07-17 23:59   ` Chris Mason
2025-07-18 18:02     ` Steven Rostedt
2025-07-02 11:49 ` [PATCH v2 08/12] sched: Re-arrange __ttwu_queue_wakelist() Peter Zijlstra
2025-07-02 11:49 ` [PATCH v2 09/12] sched: Clean up ttwu comments Peter Zijlstra
2025-07-02 11:49 ` [PATCH v2 10/12] sched: Use lock guard in sched_ttwu_pending() Peter Zijlstra
2025-07-10 16:51   ` Vincent Guittot
2025-07-02 11:49 ` [PATCH v2 11/12] sched: Change ttwu_runnable() vs sched_delayed Peter Zijlstra
2025-07-02 11:49 ` [PATCH v2 12/12] sched: Add ttwu_queue support for delayed tasks Peter Zijlstra
2025-07-03 16:00   ` Phil Auld
2025-07-03 16:47     ` Peter Zijlstra
2025-07-03 17:11       ` Phil Auld
2025-07-14 13:57         ` Phil Auld
2025-07-04  6:13       ` K Prateek Nayak
2025-07-04  7:59         ` Peter Zijlstra
2025-07-08 12:44   ` Dietmar Eggemann
2025-07-08 18:57     ` Peter Zijlstra
2025-07-08 21:02     ` Peter Zijlstra
2025-07-23  5:42   ` Shrikanth Hegde
2025-07-02 15:27 ` [PATCH v2 00/12] sched: Address schbench regression Chris Mason
2025-07-07  9:05 ` Shrikanth Hegde
2025-07-07  9:11   ` Peter Zijlstra
2025-07-07  9:38     ` Shrikanth Hegde
2025-07-16 13:46       ` Phil Auld
2025-07-17 17:25         ` Phil Auld
2025-07-07 18:19   ` Shrikanth Hegde
2025-07-08 19:02     ` Peter Zijlstra
2025-07-09 16:46       ` Shrikanth Hegde
2025-07-14 17:54       ` Shrikanth Hegde
2025-07-21 19:37       ` Shrikanth Hegde
2025-07-22 20:20         ` Chris Mason
2025-07-24 18:23           ` Chris Mason
2025-07-08 15:09   ` Chris Mason
2025-07-08 17:29     ` Shrikanth Hegde
2025-07-17 13:04 ` Beata Michalska
2025-07-17 16:57   ` Beata Michalska

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=20250702121158.817814031@infradead.org \
    --to=peterz@infradead.org \
    --cc=bsegall@google.com \
    --cc=clm@meta.com \
    --cc=dietmar.eggemann@arm.com \
    --cc=juri.lelli@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mgorman@suse.de \
    --cc=mingo@redhat.com \
    --cc=rostedt@goodmis.org \
    --cc=vincent.guittot@linaro.org \
    --cc=vschneid@redhat.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

all inboxes | Powered by JetHome®