| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <20180907152923.oxsmcqciez4yhmkk@queper01-lin>
Date: Fri, 7 Sep 2018 16:29:25 +0100
From: Quentin Perret <quentin.perret@....com>
To: "Rafael J. Wysocki" <rjw@...ysocki.net>
Cc: "Rafael J. Wysocki" <rafael@...nel.org>,
Peter Zijlstra <peterz@...radead.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
Linux PM <linux-pm@...r.kernel.org>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
Ingo Molnar <mingo@...hat.com>,
Dietmar Eggemann <dietmar.eggemann@....com>,
Morten Rasmussen <morten.rasmussen@....com>,
Chris Redpath <chris.redpath@....com>,
Patrick Bellasi <patrick.bellasi@....com>,
Valentin Schneider <valentin.schneider@....com>,
Vincent Guittot <vincent.guittot@...aro.org>,
Thara Gopinath <thara.gopinath@...aro.org>,
Viresh Kumar <viresh.kumar@...aro.org>,
Todd Kjos <tkjos@...gle.com>,
Joel Fernandes <joel@...lfernandes.org>,
Steve Muckle <smuckle@...gle.com>, adharmap@...eaurora.org,
Saravana Kannan <skannan@...eaurora.org>,
Pavan Kondeti <pkondeti@...eaurora.org>,
Juri Lelli <juri.lelli@...hat.com>,
Eduardo Valentin <edubezval@...il.com>,
Srinivas Pandruvada <srinivas.pandruvada@...ux.intel.com>,
currojerez@...eup.net, Javi Merino <javi.merino@...nel.org>
Subject: Re: [PATCH v6 13/14] sched/topology: Make Energy Aware Scheduling
depend on schedutil
On Friday 07 Sep 2018 at 10:52:01 (+0200), Rafael J. Wysocki wrote:
> On Thursday, September 6, 2018 4:38:44 PM CEST Quentin Perret wrote:
> > Hi Rafael,
> >
> > On Thursday 06 Sep 2018 at 11:18:55 (+0200), Rafael J. Wysocki wrote:
> > > I'm not a particular fan of notifiers to be honest and you don't need
> > > to add an extra chain just in order to be able to register a callback
> > > from a single user.
> >
> > Right. I agree there are alternatives to using notifiers. I used them
> > because they're existing infrastructure, and because they let me do what
> > I want without too much troubles, which are two important points.
> >
> > > That can be achieved with a single callback
> > > pointer too, but also you could just call a function exported by the
> > > scheduler directly from where in the cpufreq code it needs to be
> > > called.
> >
> > Are you thinking about something comparable to what is done in
> > cpufreq_add_update_util_hook() (kernel/sched/cpufreq.c) for example ?
> > That would probably have the same drawback as my current implementation,
> > that is that the scheduler is notified of _all_ governor changes, not
> > only changes to/from sugov although this is the only thing we care about
> > for EAS.
>
> Well, why don't you implement it as something like "if the governor changes
> from sugov to something else (or the other way around), call this function
> from the scheduler"?
I just gave it a try and ended up with the diff below. It's basically
the exact same patch with a direct function call instead of a notifier.
(I also tried the sugov_start/stop thing I keep mentioning but it is
more complex, so let's see if the simplest solution could work first).
What do you think ?
----------8<----------
diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index b0dfd3222013..6300668ac67a 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -25,6 +25,7 @@
#include <linux/kernel_stat.h>
#include <linux/module.h>
#include <linux/mutex.h>
+#include <linux/sched/cpufreq.h>
#include <linux/slab.h>
#include <linux/suspend.h>
#include <linux/syscore_ops.h>
@@ -2271,6 +2272,7 @@ static int cpufreq_set_policy(struct cpufreq_policy *policy,
ret = cpufreq_start_governor(policy);
if (!ret) {
pr_debug("cpufreq: governor change\n");
+ sched_governor_change(policy, old_gov);
return 0;
}
cpufreq_exit_governor(policy);
diff --git a/include/linux/sched/cpufreq.h b/include/linux/sched/cpufreq.h
index afa940cd50dc..33b77eed8a41 100644
--- a/include/linux/sched/cpufreq.h
+++ b/include/linux/sched/cpufreq.h
@@ -2,6 +2,7 @@
#ifndef _LINUX_SCHED_CPUFREQ_H
#define _LINUX_SCHED_CPUFREQ_H
+#include <linux/cpufreq.h>
#include <linux/types.h>
/*
@@ -28,4 +29,12 @@ static inline unsigned long map_util_freq(unsigned long util,
}
#endif /* CONFIG_CPU_FREQ */
+#if defined(CONFIG_ENERGY_MODEL) && defined(CONFIG_CPU_FREQ_GOV_SCHEDUTIL)
+void sched_governor_change(struct cpufreq_policy *policy,
+ struct cpufreq_governor *old_gov);
+#else
+static inline void sched_governor_change(struct cpufreq_policy *policy,
+ struct cpufreq_governor *old_gov) { }
+#endif
+
#endif /* _LINUX_SCHED_CPUFREQ_H */
diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
index 8356cb0072a6..2ff40b2a8ba0 100644
--- a/kernel/sched/cpufreq_schedutil.c
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -632,7 +632,7 @@ static struct kobj_type sugov_tunables_ktype = {
/********************** cpufreq governor interface *********************/
-static struct cpufreq_governor schedutil_gov;
+struct cpufreq_governor schedutil_gov;
static struct sugov_policy *sugov_policy_alloc(struct cpufreq_policy *policy)
{
@@ -891,7 +891,7 @@ static void sugov_limits(struct cpufreq_policy *policy)
sg_policy->need_freq_update = true;
}
-static struct cpufreq_governor schedutil_gov = {
+struct cpufreq_governor schedutil_gov = {
.name = "schedutil",
.owner = THIS_MODULE,
.dynamic_switching = true,
@@ -914,3 +914,32 @@ static int __init sugov_register(void)
return cpufreq_register_governor(&schedutil_gov);
}
fs_initcall(sugov_register);
+
+#ifdef CONFIG_ENERGY_MODEL
+extern bool sched_energy_update;
+static DEFINE_MUTEX(rebuild_sd_mutex);
+
+static void rebuild_sd_workfn(struct work_struct *work)
+{
+ mutex_lock(&rebuild_sd_mutex);
+ sched_energy_update = true;
+ rebuild_sched_domains();
+ sched_energy_update = false;
+ mutex_unlock(&rebuild_sd_mutex);
+}
+static DECLARE_WORK(rebuild_sd_work, rebuild_sd_workfn);
+
+/*
+ * EAS shouldn't be attempted without sugov, so rebuild the sched_domains
+ * on governor changes to make sure the scheduler knows about it.
+ */
+void sched_governor_change(struct cpufreq_policy *policy,
+ struct cpufreq_governor *old_gov)
+{
+ if (old_gov == &schedutil_gov || policy->governor == &schedutil_gov) {
+ /* Sched domains cannot be rebuilt directly from this context */
+ schedule_work(&rebuild_sd_work);
+ }
+
+}
+#endif
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index e594a854977f..915766600568 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -2265,10 +2265,8 @@ unsigned long scale_irq_capacity(unsigned long util, unsigned long irq, unsigned
}
#endif
-#ifdef CONFIG_SMP
-#ifdef CONFIG_ENERGY_MODEL
+#if defined(CONFIG_ENERGY_MODEL) && defined(CONFIG_CPU_FREQ_GOV_SCHEDUTIL)
#define perf_domain_span(pd) (to_cpumask(((pd)->obj->cpus)))
#else
#define perf_domain_span(pd) NULL
#endif
-#endif
diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c
index ae329447a082..781f3eba840e 100644
--- a/kernel/sched/topology.c
+++ b/kernel/sched/topology.c
@@ -209,7 +209,9 @@ sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent)
*/
DEFINE_STATIC_KEY_FALSE(sched_energy_present);
-#ifdef CONFIG_ENERGY_MODEL
+#if defined(CONFIG_ENERGY_MODEL) && defined(CONFIG_CPU_FREQ_GOV_SCHEDUTIL)
+bool sched_energy_update;
+
static void free_pd(struct perf_domain *pd)
{
struct perf_domain *tmp;
@@ -297,12 +299,15 @@ static void destroy_perf_domain_rcu(struct rcu_head *rp)
*/
#define EM_MAX_COMPLEXITY 2048
+extern struct cpufreq_governor schedutil_gov;
static void build_perf_domains(const struct cpumask *cpu_map)
{
int i, nr_pd = 0, nr_cs = 0, nr_cpus = cpumask_weight(cpu_map);
struct perf_domain *pd = NULL, *tmp;
int cpu = cpumask_first(cpu_map);
struct root_domain *rd = cpu_rq(cpu)->rd;
+ struct cpufreq_policy *policy;
+ struct cpufreq_governor *gov;
/* EAS is enabled for asymmetric CPU capacity topologies. */
if (!per_cpu(sd_asym_cpucapacity, cpu)) {
@@ -318,6 +323,15 @@ static void build_perf_domains(const struct cpumask *cpu_map)
if (find_pd(pd, i))
continue;
+ /* Do not attempt EAS if schedutil is not being used. */
+ policy = cpufreq_cpu_get(i);
+ if (!policy)
+ goto free;
+ gov = policy->governor;
+ cpufreq_cpu_put(policy);
+ if (gov != &schedutil_gov)
+ goto free;
+
/* Create the new pd and add it to the local list. */
tmp = pd_init(i);
if (!tmp)
@@ -389,7 +403,7 @@ static void sched_energy_start(int ndoms_new, cpumask_var_t doms_new[])
}
#else
static void free_pd(struct perf_domain *pd) { }
-#endif /* CONFIG_ENERGY_MODEL */
+#endif /* CONFIG_ENERGY_MODEL && CONFIG_CPU_FREQ_GOV_SCHEDUTIL*/
static void free_rootdomain(struct rcu_head *rcu)
{
@@ -2190,10 +2204,10 @@ void partition_sched_domains(int ndoms_new, cpumask_var_t doms_new[],
;
}
-#ifdef CONFIG_ENERGY_MODEL
+#if defined(CONFIG_ENERGY_MODEL) && defined(CONFIG_CPU_FREQ_GOV_SCHEDUTIL)
/* Build perf. domains: */
for (i = 0; i < ndoms_new; i++) {
- for (j = 0; j < n; j++) {
+ for (j = 0; j < n && !sched_energy_update; j++) {
if (cpumask_equal(doms_new[i], doms_cur[j]) &&
cpu_rq(cpumask_first(doms_cur[j]))->rd->pd)
goto match3;
Powered by blists - more mailing lists