Re: [PATCH v5 5/7] x86/pti: Do not enable PTI on processors which are not vulnerable to Meltdown

[Dave Hansen]

On 01/25/2018 10:10 AM, Dave Hansen wrote:
> On 01/25/2018 08:14 AM, David Woodhouse wrote:
>> +static bool __init early_cpu_vulnerable_meltdown(struct cpuinfo_x86 *c)
>> +{
>> +	u64 ia32_cap = 0;
>> +
>> +	if (x86_match_cpu(cpu_no_meltdown))
>> +                return false;
>> +
>> +	if (cpu_has(c, X86_FEATURE_ARCH_CAPABILITIES))
>> +		rdmsrl(MSR_IA32_ARCH_CAPABILITIES, ia32_cap);
>> +
>> +	/* Rogue Data Cache Load? No! */
>> +	if (ia32_cap & ARCH_CAP_RDCL_NO)
>> +		return false;
>> +
>> +	return true;
>> +}
> 
> Feel free to add my ack on this.  It looks fine to me.  I'll test and
> submit any necessary fixes once I actually get a system that has this bit.

Well, that was fast.  A system supporting the RDCL_NO bit booted this
code.  It reported not being affected by meltdown:

foo:/sys/devices/system/cpu/vulnerabilities/ # grep . *
meltdown:Not affected
spectre_v1:Vulnerable
spectre_v2:Vulnerable: Minimal generic ASM retpoline
foo:/sys/devices/system/cpu/vulnerabilities/ #