[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <3361688.44csPzL39Z@rjwysocki.net>
Date: Thu, 05 Jun 2025 17:05:24 +0200
From: "Rafael J. Wysocki" <rjw@...ysocki.net>
To: x86 Maintainers <x86@...nel.org>, Linux PM <linux-pm@...r.kernel.org>
Cc: LKML <linux-kernel@...r.kernel.org>, Len Brown <lenb@...nel.org>,
Peter Zijlstra <peterz@...radead.org>, Thomas Gleixner <tglx@...utronix.de>,
Dave Hansen <dave.hansen@...ux.intel.com>,
Artem Bityutskiy <artem.bityutskiy@...ux.intel.com>,
"Gautham R. Shenoy" <gautham.shenoy@....com>, Ingo Molnar <mingo@...hat.com>,
Borislav Petkov <bp@...en8.de>, Linux ACPI <linux-acpi@...r.kernel.org>
Subject: [PATCH v1 2/5] x86/smp: PM/hibernate: Split arch_resume_nosmt()
From: Rafael J. Wysocki <rafael.j.wysocki@...el.com>
Move the inner part of the arch_resume_nosmt() code into a separate
function called arch_cpu_rescan_dead_smt_siblings(), so it can be
used in other places where "dead" SMT siblings may need to be taken
online and offline again in order to get into deep idle states.
No intentional functional impact.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@...el.com>
---
arch/x86/kernel/smp.c | 23 +++++++++++++++++++++++
arch/x86/power/hibernate.c | 17 +++++------------
include/linux/cpu.h | 1 +
3 files changed, 29 insertions(+), 12 deletions(-)
--- a/arch/x86/kernel/smp.c
+++ b/arch/x86/kernel/smp.c
@@ -299,3 +299,26 @@
.send_call_func_single_ipi = native_send_call_func_single_ipi,
};
EXPORT_SYMBOL_GPL(smp_ops);
+
+int arch_cpu_rescan_dead_smt_siblings(void)
+{
+ enum cpuhp_smt_control old = cpu_smt_control;
+ int ret;
+
+ /*
+ * If SMT has been disabled and SMT siblings are in HLT, bring them back
+ * online and offline them again so that they end up in MWAIT proper.
+ *
+ * Called with hotplug enabled.
+ */
+ if (old != CPU_SMT_DISABLED && old != CPU_SMT_FORCE_DISABLED)
+ return 0;
+
+ ret = cpuhp_smt_enable();
+ if (ret)
+ return ret;
+
+ ret = cpuhp_smt_disable(old);
+
+ return ret;
+}
--- a/arch/x86/power/hibernate.c
+++ b/arch/x86/power/hibernate.c
@@ -188,7 +188,8 @@
int arch_resume_nosmt(void)
{
- int ret = 0;
+ int ret;
+
/*
* We reached this while coming out of hibernation. This means
* that SMT siblings are sleeping in hlt, as mwait is not safe
@@ -202,18 +203,10 @@
* Called with hotplug disabled.
*/
cpu_hotplug_enable();
- if (cpu_smt_control == CPU_SMT_DISABLED ||
- cpu_smt_control == CPU_SMT_FORCE_DISABLED) {
- enum cpuhp_smt_control old = cpu_smt_control;
- ret = cpuhp_smt_enable();
- if (ret)
- goto out;
- ret = cpuhp_smt_disable(old);
- if (ret)
- goto out;
- }
-out:
+ ret = arch_cpu_rescan_dead_smt_siblings();
+
cpu_hotplug_disable();
+
return ret;
}
--- a/include/linux/cpu.h
+++ b/include/linux/cpu.h
@@ -172,6 +172,7 @@
void arch_tick_broadcast_enter(void);
void arch_tick_broadcast_exit(void);
void __noreturn arch_cpu_idle_dead(void);
+int arch_cpu_rescan_dead_smt_siblings(void);
#ifdef CONFIG_ARCH_HAS_CPU_FINALIZE_INIT
void arch_cpu_finalize_init(void);
Powered by blists - more mailing lists