Re: [PATCH v3] cpu/hotplug: Fix NULL kobject warning in cpuhp_smt_enable()

[Jinjie Ruan]

On 6/2/2026 7:09 PM, Will Deacon wrote:
> On Wed, May 20, 2026 at 10:20:23AM +0800, Jinjie Ruan wrote:
>> On arm64, when booting with `maxcpus` greater than the number of present
>> CPUs (e.g., QEMU -smp cpus=4,maxcpus=8), some CPUs are marked as 'present'
>> but have not yet been registered via register_cpu(). Consequently,
>> the per-cpu device objects for these CPUs are not yet initialized.
>>
>> In cpuhp_smt_enable(), the code iterates over all present CPUs. Calling
>> _cpu_up() for these unregistered CPUs eventually leads to
>> sysfs_create_group() being called with a NULL kobject (or a kobject
>> without a directory), triggering the following warning in
>> fs/sysfs/group.c:
>>
>> 	if (WARN_ON(!kobj || (!update && !kobj->sd)))
>> 		return -EINVAL;
>>
>> When booting with ACPI, arm64 smp_prepare_cpus() currently sets all
>> enumerated CPUs as "present" regardless of their status in the MADT. This
>> causes issues with SMT hotplug control. For instance, with QEMU's
>> "-smp 4,maxcpus=8" configuration, the MADT GICC entries are populated as
>> follows: the first four CPUs are marked Enabled while the remaining four
>> are marked Online Capable to support potential hot-plugging.
>>
>> Fix this by:
>>
>> 1. When booting with ACPI, checking the ACPI_MADT_ENABLED flag in the GICC
>>    entry before calling set_cpu_present() during SMP initialization.
>>
>> 2. Properly managing the present mask in acpi_map_cpu() and
>>    acpi_unmap_cpu() to support actual CPU hotplug events, This aligns with
>>    other architectures like x86 and LoongArch.
>>
>> 3. Update the arm64 CPU hotplug documentation to no longer state that all
>>    online-capable vCPUs are marked as present by the kernel at boot time.
>>
>> This ensures that only physically available or explicitly enabled CPUs
>> are in the present mask, keeping the SMT control logic consistent with
>> the actual hardware state.
> 
> Please can you check the Sashiko review comment?
> 
> https://sashiko.dev/#/patchset/20260520022023.126670-1-ruanjinjie@xxxxxxxxxx

On arm64, arch_unregister_cpu() enforces a safety block against physical
CPU hot-removal by aborting early if cpu_present() is true but the
device is no longer physically present, thereby skipping
unregister_cpu() and leaking the sysfs device node.
If acpi_unmap_cpu() blindly clears the present bit while the sysfs
device remains registered, a subsequent hot-add attempt will see the
valid leaked device pointer, skip acpi_processor_hotadd_init() (and thus
skip acpi_map_cpu()), leaving the hot-added CPU permanently absent and
deadlocked.

Hi, Will,

what do you think about fix it like this?

--- a/arch/arm64/kernel/acpi.c
+++ b/arch/arm64/kernel/acpi.c
@@ -455,7 +455,6 @@ EXPORT_SYMBOL(acpi_map_cpu);

 int acpi_unmap_cpu(int cpu)
 {
-       set_cpu_present(cpu, false);
        return 0;
 }
 EXPORT_SYMBOL(acpi_unmap_cpu);
diff --git a/arch/arm64/kernel/smp.c b/arch/arm64/kernel/smp.c
index 5932e5b30b71..507c6d761434 100644
--- a/arch/arm64/kernel/smp.c
+++ b/arch/arm64/kernel/smp.c
@@ -554,6 +554,7 @@ void arch_unregister_cpu(int cpu)
        }

        unregister_cpu(c);
+       set_cpu_present(cpu, false);
 }
 #endif /* CONFIG_ACPI_HOTPLUG_CPU */


> 
> Cheers,
> 
> Will
>