From: Thomas Gleixner <tglx@linutronix.de>
To: speck@linutronix.de
Subject: [patch V4 01/13] sched/smt: Update sched_smt_present at runtime
Date: Wed, 20 Jun 2018 22:19:08 +0200 [thread overview]
Message-ID: <20180620201932.556507664@linutronix.de> (raw)
In-Reply-To: 20180620201907.304694346@linutronix.de
Subject: [patch V4 01/13] sched/smt: Update sched_smt_present at runtime
From: Peter Zijlstra <peterz@infradead.org>
The static key sched_smt_present is only updated at boot time when SMT
siblings have been detected. Booting with maxcpus=1 and bringing the
siblings online after boot rebuilds the scheduling domains correctly but
does not update the static key, so the SMT code is not enabled.
Let the key be updated in the scheduler CPU hotplug code to fix this.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
---
V3: Fixed changelog
kernel/sched/core.c | 30 ++++++++++++------------------
kernel/sched/fair.c | 1 +
2 files changed, 13 insertions(+), 18 deletions(-)
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -5777,6 +5777,18 @@ int sched_cpu_activate(unsigned int cpu)
struct rq *rq = cpu_rq(cpu);
struct rq_flags rf;
+#ifdef CONFIG_SCHED_SMT
+ /*
+ * The sched_smt_present static key needs to be evaluated on every
+ * hotplug event because at boot time SMT might be disabled when
+ * the number of booted CPUs is limited.
+ *
+ * If then later a sibling gets hotplugged, then the key would stay
+ * off and SMT scheduling would never be functional.
+ */
+ if (cpumask_weight(cpu_smt_mask(cpu)) > 1)
+ static_branch_enable_cpuslocked(&sched_smt_present);
+#endif
set_cpu_active(cpu, true);
if (sched_smp_initialized) {
@@ -5874,22 +5886,6 @@ int sched_cpu_dying(unsigned int cpu)
}
#endif
-#ifdef CONFIG_SCHED_SMT
-DEFINE_STATIC_KEY_FALSE(sched_smt_present);
-
-static void sched_init_smt(void)
-{
- /*
- * We've enumerated all CPUs and will assume that if any CPU
- * has SMT siblings, CPU0 will too.
- */
- if (cpumask_weight(cpu_smt_mask(0)) > 1)
- static_branch_enable(&sched_smt_present);
-}
-#else
-static inline void sched_init_smt(void) { }
-#endif
-
void __init sched_init_smp(void)
{
sched_init_numa();
@@ -5911,8 +5907,6 @@ void __init sched_init_smp(void)
init_sched_rt_class();
init_sched_dl_class();
- sched_init_smt();
-
sched_smp_initialized = true;
}
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -6238,6 +6238,7 @@ static inline int find_idlest_cpu(struct
}
#ifdef CONFIG_SCHED_SMT
+DEFINE_STATIC_KEY_FALSE(sched_smt_present);
static inline void set_idle_cores(int cpu, int val)
{
next prev parent reply other threads:[~2018-06-20 20:39 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-06-20 20:19 [patch V4 00/13] SMT control knobs Thomas Gleixner
2018-06-20 20:19 ` Thomas Gleixner [this message]
2018-06-20 20:19 ` [patch V4 02/13] x86/smp: Provide topology_is_primary_thread() Thomas Gleixner
2018-06-27 13:56 ` [MODERATED] " Josh Poimboeuf
2018-06-27 15:54 ` Thomas Gleixner
2018-06-27 16:20 ` [MODERATED] " Josh Poimboeuf
2018-06-27 16:52 ` Thomas Gleixner
2018-06-20 20:19 ` [patch V4 03/13] cpu/hotplug: Make bringup/teardown of smp threads symmetric Thomas Gleixner
2018-06-20 20:19 ` [patch V4 04/13] cpu/hotplug: Split do_cpu_down() Thomas Gleixner
2018-06-20 20:19 ` [patch V4 05/13] cpu/hotplug: Provide knobs to control SMT Thomas Gleixner
2018-06-20 23:05 ` [MODERATED] " Greg KH
2018-06-21 6:08 ` Thomas Gleixner
2018-06-20 20:19 ` [patch V4 06/13] x86/cpu: Remove the pointless CPU printout Thomas Gleixner
2018-06-20 20:19 ` [patch V4 07/13] x86/cpu/AMD: Remove the pointless detect_ht() call Thomas Gleixner
2018-06-20 20:19 ` [patch V4 08/13] x86/cpu/common: Provide detect_ht_early() Thomas Gleixner
2018-06-20 20:19 ` [patch V4 09/13] x86/cpu/topology: Provide detect_extended_topology_early() Thomas Gleixner
2018-06-20 20:19 ` [patch V4 10/13] x86/cpu/intel: Evaluate smp_num_siblings early Thomas Gleixner
2018-06-20 20:19 ` [patch V4 11/13] x86/CPU/AMD: Do not check CPUID max ext level before parsing SMP info Thomas Gleixner
2018-06-20 20:19 ` [patch V4 12/13] x86/cpu/AMD: Evaluate smp_num_siblings early Thomas Gleixner
2018-06-20 20:19 ` [patch V4 13/13] x86/apic: Ignore secondary threads if nosmt=force Thomas Gleixner
2018-06-20 21:44 ` [MODERATED] " Linus Torvalds
2018-06-28 22:13 ` Dave Hansen
2018-06-28 22:19 ` Andrew Cooper
2018-06-28 22:23 ` Luck, Tony
2018-06-28 22:29 ` Andrew Cooper
2018-06-29 8:26 ` [MODERATED] " Borislav Petkov
2018-06-29 17:01 ` Luck, Tony
2018-06-29 17:08 ` Linus Torvalds
2018-06-29 18:36 ` Thomas Gleixner
2018-06-29 8:50 ` Thomas Gleixner
2018-06-29 16:48 ` [MODERATED] " Dave Hansen
2018-06-20 20:53 ` [patch V4 00/13] SMT control knobs Thomas Gleixner
2018-06-21 11:52 ` [MODERATED] " Ingo Molnar
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=20180620201932.556507664@linutronix.de \
--to=tglx@linutronix.de \
--cc=speck@linutronix.de \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.