From: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
To: "Peter Zijlstra (Intel)" <peterz@infradead.org>,
Juri Lelli <juri.lelli@redhat.com>,
Vincent Guittot <vincent.guittot@linaro.org>
Cc: Ricardo Neri <ricardo.neri@intel.com>,
"Ravi V. Shankar" <ravi.v.shankar@intel.com>,
Ben Segall <bsegall@google.com>,
Daniel Bristot de Oliveira <bristot@redhat.com>,
Dietmar Eggemann <dietmar.eggemann@arm.com>,
Len Brown <len.brown@intel.com>, Mel Gorman <mgorman@suse.de>,
"Rafael J. Wysocki" <rafael.j.wysocki@intel.com>,
Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>,
Steven Rostedt <rostedt@goodmis.org>,
Tim Chen <tim.c.chen@linux.intel.com>,
Valentin Schneider <vschneid@redhat.com>,
x86@kernel.org,
"Joel Fernandes (Google)" <joel@joelfernandes.org>,
linux-kernel@vger.kernel.org, linux-pm@vger.kernel.org,
Ricardo Neri <ricardo.neri-calderon@linux.intel.com>,
"Tim C . Chen" <tim.c.chen@intel.com>
Subject: [PATCH v2 07/22] sched/fair: Compute IPC class scores for load balancing
Date: Mon, 28 Nov 2022 05:20:45 -0800 [thread overview]
Message-ID: <20221128132100.30253-8-ricardo.neri-calderon@linux.intel.com> (raw)
In-Reply-To: <20221128132100.30253-1-ricardo.neri-calderon@linux.intel.com>
Compute the joint total (both current and prospective) IPC class score of
a scheduling group and the local scheduling group.
These IPCC statistics are used during asym_packing load balancing. It
implies that the candidate sched group will have one fewer busy CPU after
load balancing. This observation is important for physical cores with
SMT support.
The IPCC score of scheduling groups composed of SMT siblings needs to
consider that the siblings share CPU resources. When computing the total
IPCC score of the scheduling group, divide score from each sibilng by
the number of busy siblings.
Cc: Ben Segall <bsegall@google.com>
Cc: Daniel Bristot de Oliveira <bristot@redhat.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Joel Fernandes (Google) <joel@joelfernandes.org>
Cc: Len Brown <len.brown@intel.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tim C. Chen <tim.c.chen@intel.com>
Cc: Valentin Schneider <vschneid@redhat.com>
Cc: x86@kernel.org
Cc: linux-pm@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
---
Changes since v1:
* Implemented cleanups and reworks from PeterZ. I took all his
suggestions, except the computation of the IPC score before and after
load balancing. We are computing not the average score, but the *total*.
* Check for the SD_SHARE_CPUCAPACITY to compute the throughput of the SMT
siblings of a physical core.
* Used the new interface names.
* Reworded commit message for clarity.
---
kernel/sched/fair.c | 45 +++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 45 insertions(+)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 3a1d6c50a19b..e333f9623b3a 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -8766,6 +8766,10 @@ struct sg_lb_stats {
unsigned int nr_numa_running;
unsigned int nr_preferred_running;
#endif
+#ifdef CONFIG_IPC_CLASSES
+ long ipcc_score_after; /* Prospective IPCC score after load balancing */
+ long ipcc_score_before; /* IPCC score before load balancing */
+#endif
};
/*
@@ -9140,6 +9144,38 @@ static void update_sg_lb_ipcc_stats(struct sg_lb_ipcc_stats *sgcs,
}
}
+static void update_sg_lb_stats_scores(struct sg_lb_ipcc_stats *sgcs,
+ struct sg_lb_stats *sgs,
+ struct sched_group *sg,
+ int dst_cpu)
+{
+ int busy_cpus, score_on_dst_cpu;
+ long before, after;
+
+ if (!sched_ipcc_enabled())
+ return;
+
+ busy_cpus = sgs->group_weight - sgs->idle_cpus;
+ /* No busy CPUs in the group. No tasks to move. */
+ if (!busy_cpus)
+ return;
+
+ score_on_dst_cpu = arch_get_ipcc_score(sgcs->min_ipcc, dst_cpu);
+
+ before = sgcs->sum_score;
+ after = before - sgcs->min_score;
+
+ /* SMT siblings share throughput. */
+ if (busy_cpus > 1 && sg->flags & SD_SHARE_CPUCAPACITY) {
+ before /= busy_cpus;
+ /* One sibling will become idle after load balance. */
+ after /= busy_cpus - 1;
+ }
+
+ sgs->ipcc_score_after = after + score_on_dst_cpu;
+ sgs->ipcc_score_before = before;
+}
+
#else /* CONFIG_IPC_CLASSES */
static void update_sg_lb_ipcc_stats(struct sg_lb_ipcc_stats *sgcs,
struct rq *rq)
@@ -9149,6 +9185,14 @@ static void update_sg_lb_ipcc_stats(struct sg_lb_ipcc_stats *sgcs,
static void init_rq_ipcc_stats(struct sg_lb_ipcc_stats *class_sgs)
{
}
+
+static void update_sg_lb_stats_scores(struct sg_lb_ipcc_stats *sgcs,
+ struct sg_lb_stats *sgs,
+ struct sched_group *sg,
+ int dst_cpu)
+{
+}
+
#endif /* CONFIG_IPC_CLASSES */
/**
@@ -9329,6 +9373,7 @@ static inline void update_sg_lb_stats(struct lb_env *env,
if (!local_group && env->sd->flags & SD_ASYM_PACKING &&
env->idle != CPU_NOT_IDLE && sgs->sum_h_nr_running &&
sched_asym(env, sds, sgs, group)) {
+ update_sg_lb_stats_scores(&sgcs, sgs, group, env->dst_cpu);
sgs->group_asym_packing = 1;
}
--
2.25.1
next prev parent reply other threads:[~2022-11-28 13:14 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-11-28 13:20 [PATCH v2 00/22] sched: Introduce IPC classes for load balance Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 01/22] sched/task_struct: Introduce IPC classes of tasks Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 02/22] sched: Add interfaces for IPC classes Ricardo Neri
2022-12-08 8:48 ` Ionela Voinescu
2022-12-14 0:31 ` Ricardo Neri
2022-12-14 23:15 ` Ionela Voinescu
2022-12-20 0:12 ` Ricardo Neri
2022-12-14 7:36 ` Lukasz Luba
2022-12-16 21:56 ` Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 03/22] sched/core: Initialize the IPC class of a new task Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 04/22] sched/core: Add user_tick as argument to scheduler_tick() Ricardo Neri
2022-12-07 12:21 ` Dietmar Eggemann
2022-12-12 18:47 ` Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 05/22] sched/core: Update the IPC class of the current task Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 06/22] sched/fair: Collect load-balancing stats for IPC classes Ricardo Neri
2022-12-07 17:00 ` Dietmar Eggemann
2022-12-12 21:41 ` Ricardo Neri
2022-12-08 8:50 ` Ionela Voinescu
2022-12-14 0:31 ` Ricardo Neri
2022-11-28 13:20 ` Ricardo Neri [this message]
2022-11-28 13:20 ` [PATCH v2 08/22] sched/fair: Use IPC class to pick the busiest group Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 09/22] sched/fair: Use IPC class score to select a busiest runqueue Ricardo Neri
2022-12-08 8:51 ` Ionela Voinescu
2022-12-14 0:32 ` Ricardo Neri
2022-12-14 23:16 ` Ionela Voinescu
2022-12-16 23:24 ` Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 10/22] thermal: intel: hfi: Introduce Intel Thread Director classes Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 11/22] thermal: intel: hfi: Store per-CPU IPCC scores Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 12/22] x86/cpufeatures: Add the Intel Thread Director feature definitions Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 13/22] thermal: intel: hfi: Update the IPC class of the current task Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 14/22] thermal: intel: hfi: Report the IPC class score of a CPU Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 15/22] thermal: intel: hfi: Define a default class for unclassified tasks Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 16/22] thermal: intel: hfi: Enable the Intel Thread Director Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 17/22] sched/task_struct: Add helpers for IPC classification Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 18/22] sched/core: Initialize helpers of task classification Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 19/22] thermal: intel: hfi: Implement model-specific checks for " Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 20/22] x86/cpufeatures: Add feature bit for HRESET Ricardo Neri
2022-11-28 13:20 ` [PATCH v2 21/22] x86/hreset: Configure history reset Ricardo Neri
2022-11-28 13:21 ` [PATCH v2 22/22] x86/process: Reset hardware history in context switch Ricardo Neri
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=20221128132100.30253-8-ricardo.neri-calderon@linux.intel.com \
--to=ricardo.neri-calderon@linux.intel.com \
--cc=bristot@redhat.com \
--cc=bsegall@google.com \
--cc=dietmar.eggemann@arm.com \
--cc=joel@joelfernandes.org \
--cc=juri.lelli@redhat.com \
--cc=len.brown@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=mgorman@suse.de \
--cc=peterz@infradead.org \
--cc=rafael.j.wysocki@intel.com \
--cc=ravi.v.shankar@intel.com \
--cc=ricardo.neri@intel.com \
--cc=rostedt@goodmis.org \
--cc=srinivas.pandruvada@linux.intel.com \
--cc=tim.c.chen@intel.com \
--cc=tim.c.chen@linux.intel.com \
--cc=vincent.guittot@linaro.org \
--cc=vschneid@redhat.com \
--cc=x86@kernel.org \
/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®