[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAJZ5v0hkMUP3rwSggAPNp+LAKnYBoNn4sE2e9xGR_d-aofwxAA@mail.gmail.com>
Date: Mon, 22 Sep 2025 17:42:40 +0200
From: "Rafael J. Wysocki" <rafael@...nel.org>
To: Thomas Gleixner <tglx@...utronix.de>
Cc: "Rafael J. Wysocki" <rafael@...nel.org>, LKML <linux-kernel@...r.kernel.org>,
Linux PM <linux-pm@...r.kernel.org>, Peter Zijlstra <peterz@...radead.org>,
Christian Loehle <christian.loehle@....com>, Dave Hansen <dave.hansen@...ux.intel.com>
Subject: Re: [PATCH] x86/topology: Implement topology_is_core_online() to
address SMT regression
On Sun, Sep 21, 2025 at 10:56 AM Thomas Gleixner <tglx@...utronix.de> wrote:
>
> Christian reported that commit a430c11f4015 ("intel_idle: Rescan "dead" SMT
> siblings during initialization") broke the use case in which both 'nosmt'
> and 'maxcpus' are on the kernel command line because it onlines primary
> threads, which were offline due to the maxcpus limit.
>
> The initially proposed fix to skip primary threads in the loop is
> inconsistent. While it prevents the primary thread to be onlined, it then
> onlines the corresponding hyperthread(s), which does not really make sense.
>
> The CPU iterator in cpuhp_smt_enable() contains a check which excludes all
> threads of a core, when the primary thread is offline. The default
> implementation is a NOOP and therefore not effective on x86.
>
> Implement topology_is_core_online() on x86 to address this issue. This
> makes the behaviour consistent between x86 and PowerPC.
>
> Fixes: a430c11f4015 ("intel_idle: Rescan "dead" SMT siblings during initialization")
> Fixes: f694481b1d31 ("ACPI: processor: Rescan "dead" SMT siblings during initialization")
> Reported-by: Christian Loehle <christian.loehle@....com>
> Signed-off-by: Thomas Gleixner <tglx@...utronix.de>
> Cc: stable@...r.kernel.org
> Link: https://lore.kernel.org/12740505.O9o76ZdvQC@rafael.j.wysocki
> Closes: https://lore.kernel.org/linux-pm/724616a2-6374-4ba3-8ce3-ea9c45e2ae3b@arm.com/
Reviewed-by: Rafael J. Wysocki (Intel) <rafael@...nel.org>
> ---
> arch/x86/include/asm/topology.h | 10 ++++++++++
> arch/x86/kernel/cpu/topology.c | 13 +++++++++++++
> 2 files changed, 23 insertions(+)
>
> --- a/arch/x86/include/asm/topology.h
> +++ b/arch/x86/include/asm/topology.h
> @@ -231,6 +231,16 @@ static inline bool topology_is_primary_t
> }
> #define topology_is_primary_thread topology_is_primary_thread
>
> +int topology_get_primary_thread(unsigned int cpu);
> +
> +static inline bool topology_is_core_online(unsigned int cpu)
> +{
> + int pcpu = topology_get_primary_thread(cpu);
> +
> + return pcpu >= 0 ? cpu_online(pcpu) : false;
> +}
> +#define topology_is_core_online topology_is_core_online
> +
> #else /* CONFIG_SMP */
> static inline int topology_phys_to_logical_pkg(unsigned int pkg) { return 0; }
> static inline int topology_max_smt_threads(void) { return 1; }
> --- a/arch/x86/kernel/cpu/topology.c
> +++ b/arch/x86/kernel/cpu/topology.c
> @@ -372,6 +372,19 @@ unsigned int topology_unit_count(u32 api
> return topo_unit_count(lvlid, at_level, apic_maps[which_units].map);
> }
>
> +#ifdef CONFIG_SMP
> +int topology_get_primary_thread(unsigned int cpu)
> +{
> + u32 apic_id = cpuid_to_apicid[cpu];
> +
> + /*
> + * Get the core domain level APIC id, which is the primary thread
> + * and return the CPU number assigned to it.
> + */
> + return topo_lookup_cpuid(topo_apicid(apic_id, TOPO_CORE_DOMAIN));
> +}
> +#endif
> +
> #ifdef CONFIG_ACPI_HOTPLUG_CPU
> /**
> * topology_hotplug_apic - Handle a physical hotplugged APIC after boot
Powered by blists - more mailing lists