Re: [PATCH 1/6] arm64: kpti: move check for non-vulnerable CPUs to a function

[Jeremy Linton]

Hi Julien,

Thanks for looking at this,

On 12/13/2018 03:13 AM, Julien Thierry wrote:
> Hi,
>
> On 06/12/2018 23:44, Jeremy Linton wrote:
> > From: Mian Yousaf Kaukab <ykaukab@xxxxxxx>
> >
> > Add is_meltdown_safe() which is a whitelist of known safe cores.
> >
> > Signed-off-by: Mian Yousaf Kaukab <ykaukab@xxxxxxx>
> > [Moved location of function]
> > Signed-off-by: Jeremy Linton <jeremy.linton@xxxxxxx>
> > ---
> >   arch/arm64/kernel/cpufeature.c | 16 ++++++++++++----
> >   1 file changed, 12 insertions(+), 4 deletions(-)
> >
> > diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c
> > index aec5ecb85737..242898395f68 100644
> > --- a/arch/arm64/kernel/cpufeature.c
> > +++ b/arch/arm64/kernel/cpufeature.c
> > @@ -908,8 +908,7 @@ has_useable_cnp(const struct arm64_cpu_capabilities *entry, int scope)
> >   #ifdef CONFIG_UNMAP_KERNEL_AT_EL0
> >   static int __kpti_forced; /* 0: not forced, >0: forced on, <0: forced off */
> >   
> > -static bool unmap_kernel_at_el0(const struct arm64_cpu_capabilities *entry,
> > -				int scope)
> > +static bool is_cpu_meltdown_safe(void)
> >   {
> >   	/* List of CPUs that are not vulnerable and don't need KPTI */
> >   	static const struct midr_range kpti_safe_list[] = {
> > @@ -917,6 +916,16 @@ static bool unmap_kernel_at_el0(const struct arm64_cpu_capabilities *entry,
> >   		MIDR_ALL_VERSIONS(MIDR_BRCM_VULCAN),
> >   		{ /* sentinel */ }
> >   	};
> > +	/* Don't force KPTI for CPUs that are not vulnerable */
>
> This is really a nit, but that comment would make more sense where
> is_cpu_meltdown_safe() is called since unmap_kernel_at_el0 is the one
> deciding whether to apply KPTI, is_cpu_meltdown_safe() just states
> whether the core is safe of not.

That is a good point, thanks.


> Otherwise:
>
> Reviewed-by: Julien Thierry <julien.thierry@xxxxxxx>
>
> Cheers,
>
> Julien
>
> > +	if (is_midr_in_range_list(read_cpuid_id(), kpti_safe_list))
> > +		return true;
> > +
> > +	return false;
> > +}
> > +
> > +static bool unmap_kernel_at_el0(const struct arm64_cpu_capabilities *entry,
> > +				int scope)
> > +{
> >   	char const *str = "command line option";
> >   
> >   	/*
> > @@ -940,8 +949,7 @@ static bool unmap_kernel_at_el0(const struct arm64_cpu_capabilities *entry,
> >   	if (IS_ENABLED(CONFIG_RANDOMIZE_BASE))
> >   		return true;
> >   
> > -	/* Don't force KPTI for CPUs that are not vulnerable */
> > -	if (is_midr_in_range_list(read_cpuid_id(), kpti_safe_list))
> > +	if (is_cpu_meltdown_safe())
> >   		return false;
> >   
> >   	/* Defer to CPU feature registers */