linux-pm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Stuart Hayes <stuart.w.hayes@gmail.com>
To: "Rafael J . Wysocki" <rafael@kernel.org>,
	Viresh Kumar <viresh.kumar@linaro.org>,
	Stuart Hayes <stuart.w.hayes@gmail.com>,
	Kyle Meyer <kyle.meyer@hpe.com>
Cc: linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH] cpufreq: acpi: Defer setting boost MSRs
Date: Wed,  2 Nov 2022 14:59:57 -0500	[thread overview]
Message-ID: <20221102195957.82871-1-stuart.w.hayes@gmail.com> (raw)

When acpi-cpufreq is loaded, boost is enabled on every CPU (by setting an
MSR) before the driver is registered with cpufreq.  This can be very time
consuming, because it is done with a CPU hotplug startup callback, and
cpuhp_setup_state() schedules the callback (cpufreq_boost_online()) to run
on each CPU one at a time, waiting for each to run before calling the next.

If cpufreq_register_driver() fails--if, for example, there are no ACPI
P-states present--this is wasted time.

Since cpufreq already sets up a CPU hotplug startup callback if and when
acpi-cpufreq is registered, set the boost MSRs in acpi_cpufreq_cpu_init(),
which is called by the cpufreq cpuhp callback.  This allows acpi-cpufreq to
exit quickly if it is loaded but not needed.

On one system with 192 CPUs, this patch speeds up boot by about 30 seconds.

Signed-off-by: Stuart Hayes <stuart.w.hayes@gmail.com>
---
 drivers/cpufreq/acpi-cpufreq.c | 31 +++----------------------------
 1 file changed, 3 insertions(+), 28 deletions(-)

diff --git a/drivers/cpufreq/acpi-cpufreq.c b/drivers/cpufreq/acpi-cpufreq.c
index 1bb2b90ebb21..cb167263de72 100644
--- a/drivers/cpufreq/acpi-cpufreq.c
+++ b/drivers/cpufreq/acpi-cpufreq.c
@@ -535,15 +535,6 @@ static void free_acpi_perf_data(void)
 	free_percpu(acpi_perf_data);
 }
 
-static int cpufreq_boost_online(unsigned int cpu)
-{
-	/*
-	 * On the CPU_UP path we simply keep the boost-disable flag
-	 * in sync with the current global state.
-	 */
-	return boost_set_msr(acpi_cpufreq_driver.boost_enabled);
-}
-
 static int cpufreq_boost_down_prep(unsigned int cpu)
 {
 	/*
@@ -897,6 +888,8 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy)
 	if (perf->states[0].core_frequency * 1000 != freq_table[0].frequency)
 		pr_warn(FW_WARN "P-state 0 is not max freq\n");
 
+	set_boost(policy, acpi_cpufreq_driver.boost_enabled);
+
 	return result;
 
 err_unreg:
@@ -916,6 +909,7 @@ static int acpi_cpufreq_cpu_exit(struct cpufreq_policy *policy)
 
 	pr_debug("%s\n", __func__);
 
+	cpufreq_boost_down_prep(policy->cpu);
 	policy->fast_switch_possible = false;
 	policy->driver_data = NULL;
 	acpi_processor_unregister_performance(data->acpi_perf_cpu);
@@ -972,25 +966,9 @@ static void __init acpi_cpufreq_boost_init(void)
 	acpi_cpufreq_driver.set_boost = set_boost;
 	acpi_cpufreq_driver.boost_enabled = boost_state(0);
 
-	/*
-	 * This calls the online callback on all online cpu and forces all
-	 * MSRs to the same value.
-	 */
-	ret = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "cpufreq/acpi:online",
-				cpufreq_boost_online, cpufreq_boost_down_prep);
-	if (ret < 0) {
-		pr_err("acpi_cpufreq: failed to register hotplug callbacks\n");
-		return;
-	}
 	acpi_cpufreq_online = ret;
 }
 
-static void acpi_cpufreq_boost_exit(void)
-{
-	if (acpi_cpufreq_online > 0)
-		cpuhp_remove_state_nocalls(acpi_cpufreq_online);
-}
-
 static int __init acpi_cpufreq_init(void)
 {
 	int ret;
@@ -1032,7 +1010,6 @@ static int __init acpi_cpufreq_init(void)
 	ret = cpufreq_register_driver(&acpi_cpufreq_driver);
 	if (ret) {
 		free_acpi_perf_data();
-		acpi_cpufreq_boost_exit();
 	}
 	return ret;
 }
@@ -1041,8 +1018,6 @@ static void __exit acpi_cpufreq_exit(void)
 {
 	pr_debug("%s\n", __func__);
 
-	acpi_cpufreq_boost_exit();
-
 	cpufreq_unregister_driver(&acpi_cpufreq_driver);
 
 	free_acpi_perf_data();
-- 
2.31.1


             reply	other threads:[~2022-11-02 20:01 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-11-02 19:59 Stuart Hayes [this message]
2022-11-03 18:19 ` [PATCH] cpufreq: acpi: Defer setting boost MSRs Rafael J. Wysocki
2022-12-04 18:29   ` Borislav Petkov
2022-12-04 19:20     ` stuart hayes
2022-12-05 12:43       ` Rafael J. Wysocki
2022-12-05 22:57         ` stuart hayes

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=20221102195957.82871-1-stuart.w.hayes@gmail.com \
    --to=stuart.w.hayes@gmail.com \
    --cc=kyle.meyer@hpe.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=rafael@kernel.org \
    --cc=viresh.kumar@linaro.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;
as well as URLs for NNTP newsgroup(s).