Re: [PATCH v3 2/4] LoongArch: Refactor get_acpi_id_for_cpu() related code

[Tiezhu Yang]

On 04/12/2024 12:17 PM, Huacai Chen wrote:
> Hi, Tiezhu,
>
> On Thu, Apr 11, 2024 at 9:05 AM Tiezhu Yang <yangtiezhu@xxxxxxxxxxx> wrote:
> > Currently, cpu_logical_map(cpu) is defined as __cpu_logical_map[cpu]
> > in arch/loongarch/include/asm/smp.h and __cpu_logical_map[] is defined
> > in arch/loongarch/kernel/smp.c, that is to say, cpu_logical_map(cpu) is
> > vaild only under CONFIG_SMP, the implementation of get_acpi_id_for_cpu()
> > which calls cpu_logical_map(cpu) is not suitable for the case of non-SMP,
> > so refactor get_acpi_id_for_cpu() related code to make it work well for
> > both SMP and non-SMP.
> But you implement cpu_logical_map(cpu) for non-SMP in the 4th patch, right?

The initial aim is to make get_acpi_id_for_cpu() only related with
acpi stuff and referenced the following riscv code, this patch can
be dropped if you think it is better to only define cpu_logical_map(cpu)
under !CONFIG_SMP, I am not sure whether it needs to keep the changes
in acpi.c at least.

static int acpi_parse_madt_rintc(union acpi_subtable_headers *header, 
const unsigned long end)
{
	struct acpi_madt_rintc *rintc = (struct acpi_madt_rintc *)header;
	int cpuid;

	if (!(rintc->flags & ACPI_MADT_ENABLED))
		return 0;

	cpuid = riscv_hartid_to_cpuid(rintc->hart_id);
	/*
	 * When CONFIG_SMP is disabled, mapping won't be created for
	 * all cpus.
	 * CPUs more than num_possible_cpus, will be ignored.
	 */
	if (cpuid >= 0 && cpuid < num_possible_cpus())
		cpu_madt_rintc[cpuid] = *rintc;

	return 0;
}

struct acpi_madt_rintc *acpi_cpu_get_madt_rintc(int cpu)
{
	return &cpu_madt_rintc[cpu];
}

u32 get_acpi_id_for_cpu(int cpu)
{
	return acpi_cpu_get_madt_rintc(cpu)->uid;
}

https://elixir.bootlin.com/linux/v6.9-rc3/source/arch/riscv/kernel/acpi.c#L194

Thanks,
Tiezhu