Re: [PATCH v3 01/17] x86/apic: Add new driver for Secure AVIC

[Neeraj Upadhyay]

On 4/3/2025 5:04 PM, Thomas Gleixner wrote:
> On Tue, Apr 01 2025 at 17:06, Neeraj Upadhyay wrote:
>> +static void __send_ipi_mask(const struct cpumask *mask, int vector, bool excl_self)
>> +{
>> +	unsigned long query_cpu;
>> +	unsigned long this_cpu;
>> +	unsigned long flags;
> 
> Just coalesce them into a single line: 'unsigned long a, b;'
>

Ok

 
>> +	/* x2apic MSRs are special and need a special fence: */
>> +	weak_wrmsr_fence();
>> +
>> +	local_irq_save(flags);
> 
>         guard(irqsave)();
> 

Ok

>> +	this_cpu = smp_processor_id();
>> +	for_each_cpu(query_cpu, mask) {
>> +		if (excl_self && this_cpu == query_cpu)
>> +			continue;
>> +		__x2apic_send_IPI_dest(per_cpu(x86_cpu_to_apicid, query_cpu),
>> +				       vector, APIC_DEST_PHYSICAL);
>> +	}
>> +	local_irq_restore(flags);
>> +}
> 
>> +static int x2apic_savic_probe(void)
>> +{
>> +	if (!cc_platform_has(CC_ATTR_SNP_SECURE_AVIC))
>> +		return 0;
>> +
>> +	if (!x2apic_mode) {
>> +		pr_err("Secure AVIC enabled in non x2APIC mode\n");
>> +		snp_abort();
> 
> Why does this return 1?
> 

snp_abort() would terminate guest execution. So, code below it is not
reachable if we reach here. I will add a comment here.



- Neeraj

>> +	}
>> +
>> +	return 1;
>> +}