Re: [RFC PATCH v9 04/27] x86/cet: Add control-protection fault handler

[Dave Hansen]

On 2/5/20 10:19 AM, Yu-cheng Yu wrote:
> diff --git a/arch/x86/kernel/idt.c b/arch/x86/kernel/idt.c
> index 87ef69a72c52..8ed406f469e7 100644
> --- a/arch/x86/kernel/idt.c
> +++ b/arch/x86/kernel/idt.c
> @@ -102,6 +102,10 @@ static const __initconst struct idt_data def_idts[] = {
>  #elif defined(CONFIG_X86_32)
>  	SYSG(IA32_SYSCALL_VECTOR,	entry_INT80_32),
>  #endif
> +
> +#ifdef CONFIG_X86_64
> +	INTG(X86_TRAP_CP,		control_protection),
> +#endif
>  };

This patch in particular appears to have all of its code unconditionally
compiled in.  That's in contrast to things that have Kconfig options, like:

#ifdef CONFIG_X86_MCE
        INTG(X86_TRAP_MC,               &machine_check),
#endif

or:

#ifdef CONFIG_X86_THERMAL_VECTOR
        INTG(THERMAL_APIC_VECTOR,       thermal_interrupt),
#endif

Is there a reason this code is always compiled in on 64-bit even when
the config option is off?