On Thu, Apr 18, 2024 at 3:57 PM Jonathan Cameron <Jonathan.Cameron@xxxxxxxxxx> wrote: > > From: James Morse <james.morse@xxxxxxx> > > The arm64 specific arch_register_cpu() call may defer CPU registration > until the ACPI interpreter is available and the _STA method can > be evaluated. > > If this occurs, then a second attempt is made in > acpi_processor_get_info(). Note that the arm64 specific call has > not yet been added so for now this will be called for the original > hotplug case. > > For architectures that do not defer until the ACPI Processor > driver loads (e.g. x86), for initially present CPUs there will > already be a CPU device. If present do not try to register again. > > Systems can still be booted with 'acpi=off', or not include an > ACPI description at all as in these cases arch_register_cpu() > will not have deferred registration when first called. > > This moves the CPU register logic back to a subsys_initcall(), > while the memory nodes will have been registered earlier. > Note this is where the call was prior to the cleanup series so > there should be no side effects of moving it back again for this > specific case. > > [PATCH 00/21] Initial cleanups for vCPU HP. > https://lore.kernel.org/all/ZVyz%2FVe5pPu8AWoA@xxxxxxxxxxxxxxxxxxxxx/ > commit 5b95f94c3b9f ("x86/topology: Switch over to GENERIC_CPU_DEVICES") > > Signed-off-by: James Morse <james.morse@xxxxxxx> > Reviewed-by: Gavin Shan <gshan@xxxxxxxxxx> > Tested-by: Miguel Luis <miguel.luis@xxxxxxxxxx> > Tested-by: Vishnu Pajjuri <vishnu@xxxxxxxxxxxxxxxxxxxxxx> > Tested-by: Jianyong Wu <jianyong.wu@xxxxxxx> > Signed-off-by: Russell King (Oracle) <rmk+kernel@xxxxxxxxxxxxxxx> > Co-developed-by: Jonathan Cameron <Jonathan.Cameron@xxxxxxxxxx> > Signed-off-by: Joanthan Cameron <Jonathan.Cameron@xxxxxxxxxx> > --- > v7: Simplify the logic on whether to hotadd the CPU. > This path can only be reached either for coldplug in which > case all we care about is has register_cpu() already been > called (identifying deferred), or hotplug in which case > whether register_cpu() has been called is also sufficient. > Checks on _STA related elements or the validity of the ID > are no longer necessary here due to similar checks having > moved elsewhere in the path. > v6: Squash the two paths for conventional CPU Hotplug and arm64 > vCPU HP. > --- > drivers/acpi/acpi_processor.c | 14 +++++++------- > 1 file changed, 7 insertions(+), 7 deletions(-) > > diff --git a/drivers/acpi/acpi_processor.c b/drivers/acpi/acpi_processor.c > index 127ae8dcb787..4e65011e706c 100644 > --- a/drivers/acpi/acpi_processor.c > +++ b/drivers/acpi/acpi_processor.c > @@ -350,14 +350,14 @@ static int acpi_processor_get_info(struct acpi_device *device) > } > > /* > - * Extra Processor objects may be enumerated on MP systems with > - * less than the max # of CPUs. They should be ignored _iff > - * they are physically not present. > - * > - * NOTE: Even if the processor has a cpuid, it may not be present > - * because cpuid <-> apicid mapping is persistent now. > + * This code is not called unless we know the CPU is present and > + * enabled. The two paths are: > + * a) Initially present CPUs on architectures that do not defer > + * their arch_register_cpu() calls until this point. > + * b) Hotplugged CPUs (enabled bit in _STA has transitioned from not > + * enabled to enabled) > */ > - if (invalid_logical_cpuid(pr->id) || !cpu_present(pr->id)) { > + if (!get_cpu_device(pr->id)) { > ret = acpi_processor_hotadd_init(pr, device); Yes, this is what I thought it should boil down to, so Acked-by: Rafael J. Wysocki <rafael.j.wysocki@xxxxxxxxx> > > if (ret) > -- > 2.39.2 >
