From: Peter Zijlstra <peterz@infradead.org>
To: Yicong Yang <yangyicong@huawei.com>
Cc: mingo@redhat.com, juri.lelli@redhat.com,
vincent.guittot@linaro.org, dietmar.eggemann@arm.com,
tim.c.chen@linux.intel.com, yu.c.chen@intel.com,
gautham.shenoy@amd.com, mgorman@suse.de, vschneid@redhat.com,
linux-kernel@vger.kernel.org,
linux-arm-kernel@lists.infradead.org, rostedt@goodmis.org,
bsegall@google.com, bristot@redhat.com, prime.zeng@huawei.com,
yangyicong@hisilicon.com, jonathan.cameron@huawei.com,
ego@linux.vnet.ibm.com, srikar@linux.vnet.ibm.com,
linuxarm@huawei.com, 21cnbao@gmail.com, kprateek.nayak@amd.com,
wuyun.abel@bytedance.com
Subject: [PATCH] sched/fair: Remove SIS_PROP
Date: Fri, 20 Oct 2023 15:43:37 +0200 [thread overview]
Message-ID: <20231020134337.GD33965@noisy.programming.kicks-ass.net> (raw)
In-Reply-To: <20231019033323.54147-1-yangyicong@huawei.com>
Since this had me looking at all that code, I did the below.
Holler if there's objections etc..
---
Subject: sched/fair: Remove SIS_PROP
From: Peter Zijlstra <peterz@infradead.org>
Date: Fri Oct 20 12:35:33 CEST 2023
SIS_UTIL seems to work well, lets remove the old thing.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
include/linux/sched/topology.h | 2 -
kernel/sched/core.c | 5 ----
kernel/sched/fair.c | 48 -----------------------------------------
kernel/sched/features.h | 1
kernel/sched/sched.h | 3 --
5 files changed, 59 deletions(-)
--- a/include/linux/sched/topology.h
+++ b/include/linux/sched/topology.h
@@ -109,8 +109,6 @@ struct sched_domain {
u64 max_newidle_lb_cost;
unsigned long last_decay_max_lb_cost;
- u64 avg_scan_cost; /* select_idle_sibling */
-
#ifdef CONFIG_SCHEDSTATS
/* load_balance() stats */
unsigned int lb_count[CPU_MAX_IDLE_TYPES];
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -3792,9 +3792,6 @@ ttwu_do_activate(struct rq *rq, struct t
if (rq->avg_idle > max)
rq->avg_idle = max;
- rq->wake_stamp = jiffies;
- rq->wake_avg_idle = rq->avg_idle / 2;
-
rq->idle_stamp = 0;
}
#endif
@@ -9991,8 +9988,6 @@ void __init sched_init(void)
rq->online = 0;
rq->idle_stamp = 0;
rq->avg_idle = 2*sysctl_sched_migration_cost;
- rq->wake_stamp = jiffies;
- rq->wake_avg_idle = rq->avg_idle;
rq->max_idle_balance_cost = sysctl_sched_migration_cost;
INIT_LIST_HEAD(&rq->cfs_tasks);
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -7209,45 +7209,9 @@ static int select_idle_cpu(struct task_s
struct cpumask *cpus = this_cpu_cpumask_var_ptr(select_rq_mask);
int i, cpu, idle_cpu = -1, nr = INT_MAX;
struct sched_domain_shared *sd_share;
- struct rq *this_rq = this_rq();
- int this = smp_processor_id();
- struct sched_domain *this_sd = NULL;
- u64 time = 0;
cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr);
- if (sched_feat(SIS_PROP) && !has_idle_core) {
- u64 avg_cost, avg_idle, span_avg;
- unsigned long now = jiffies;
-
- this_sd = rcu_dereference(*this_cpu_ptr(&sd_llc));
- if (!this_sd)
- return -1;
-
- /*
- * If we're busy, the assumption that the last idle period
- * predicts the future is flawed; age away the remaining
- * predicted idle time.
- */
- if (unlikely(this_rq->wake_stamp < now)) {
- while (this_rq->wake_stamp < now && this_rq->wake_avg_idle) {
- this_rq->wake_stamp++;
- this_rq->wake_avg_idle >>= 1;
- }
- }
-
- avg_idle = this_rq->wake_avg_idle;
- avg_cost = this_sd->avg_scan_cost + 1;
-
- span_avg = sd->span_weight * avg_idle;
- if (span_avg > 4*avg_cost)
- nr = div_u64(span_avg, avg_cost);
- else
- nr = 4;
-
- time = cpu_clock(this);
- }
-
if (sched_feat(SIS_UTIL)) {
sd_share = rcu_dereference(per_cpu(sd_llc_shared, target));
if (sd_share) {
@@ -7301,18 +7265,6 @@ static int select_idle_cpu(struct task_s
if (has_idle_core)
set_idle_cores(target, false);
- if (sched_feat(SIS_PROP) && this_sd && !has_idle_core) {
- time = cpu_clock(this) - time;
-
- /*
- * Account for the scan cost of wakeups against the average
- * idle time.
- */
- this_rq->wake_avg_idle -= min(this_rq->wake_avg_idle, time);
-
- update_avg(&this_sd->avg_scan_cost, time);
- }
-
return idle_cpu;
}
--- a/kernel/sched/features.h
+++ b/kernel/sched/features.h
@@ -49,7 +49,6 @@ SCHED_FEAT(TTWU_QUEUE, true)
/*
* When doing wakeups, attempt to limit superfluous scans of the LLC domain.
*/
-SCHED_FEAT(SIS_PROP, false)
SCHED_FEAT(SIS_UTIL, true)
/*
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -1059,9 +1059,6 @@ struct rq {
u64 idle_stamp;
u64 avg_idle;
- unsigned long wake_stamp;
- u64 wake_avg_idle;
-
/* This is used to determine avg_idle's max value */
u64 max_idle_balance_cost;
next prev parent reply other threads:[~2023-10-20 13:44 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-19 3:33 [PATCH v11 0/3] sched/fair: Scan cluster before scanning LLC in wake-up path Yicong Yang
2023-10-19 3:33 ` [PATCH v11 1/3] sched: Add cpus_share_resources API Yicong Yang
2023-10-24 8:52 ` [tip: sched/core] " tip-bot2 for Barry Song
2023-10-19 3:33 ` [PATCH v11 2/3] sched/fair: Scan cluster before scanning LLC in wake-up path Yicong Yang
2023-10-20 13:41 ` Peter Zijlstra
2023-10-24 8:52 ` [tip: sched/core] " tip-bot2 for Barry Song
2023-10-19 3:33 ` [PATCH v11 3/3] sched/fair: Use candidate prev/recent_used CPU if scanning failed for cluster wakeup Yicong Yang
2023-10-19 12:08 ` Vincent Guittot
2023-10-20 7:36 ` Chen Yu
2023-10-24 8:52 ` [tip: sched/core] " tip-bot2 for Yicong Yang
2023-10-20 13:43 ` Peter Zijlstra [this message]
2023-10-20 14:23 ` [PATCH] sched/fair: Remove SIS_PROP Vincent Guittot
2023-10-23 3:55 ` Yicong Yang
2023-10-23 8:42 ` Mel Gorman
2023-10-24 8:52 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
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=20231020134337.GD33965@noisy.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=21cnbao@gmail.com \
--cc=bristot@redhat.com \
--cc=bsegall@google.com \
--cc=dietmar.eggemann@arm.com \
--cc=ego@linux.vnet.ibm.com \
--cc=gautham.shenoy@amd.com \
--cc=jonathan.cameron@huawei.com \
--cc=juri.lelli@redhat.com \
--cc=kprateek.nayak@amd.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linuxarm@huawei.com \
--cc=mgorman@suse.de \
--cc=mingo@redhat.com \
--cc=prime.zeng@huawei.com \
--cc=rostedt@goodmis.org \
--cc=srikar@linux.vnet.ibm.com \
--cc=tim.c.chen@linux.intel.com \
--cc=vincent.guittot@linaro.org \
--cc=vschneid@redhat.com \
--cc=wuyun.abel@bytedance.com \
--cc=yangyicong@hisilicon.com \
--cc=yangyicong@huawei.com \
--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