Re: [HELPER PATCH 1] sysfs: Provide write method for paravirt

[Shrikanth Hegde <sshegde@linux.ibm.com>]

Hi Greg.

On 11/19/25 1:12 PM, Greg KH wrote:
> On Wed, Nov 19, 2025 at 11:50:59AM +0530, Shrikanth Hegde wrote:
>> This is helper patch which could be used to set the range of CPUs as
>> paravirt. One could make use of this for quick testing of this infra
>> instead of writing arch specific code.
>>
>> This is currently not meant be merged, since paravirt sysfs file is meant
>> to be Read-Only.
>>
>> echo 100-200,600-700 >  /sys/devices/system/cpu/paravirt
>> cat /sys/devices/system/cpu/paravirt
>> 100-200,600-700
>>
>> echo > /sys/devices/system/cpu/paravirt
>> cat /sys/devices/system/cpu/paravirt
>>
>> Signed-off-by: Shrikanth Hegde <sshegde@linux.ibm.com>
>> ---
>>   drivers/base/cpu.c | 48 ++++++++++++++++++++++++++++++++++++++++++++--
>>   1 file changed, 46 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c
>> index 59ceae217b22..043e4f4ce1a9 100644
>> --- a/drivers/base/cpu.c
>> +++ b/drivers/base/cpu.c
>> @@ -375,12 +375,57 @@ static int cpu_uevent(const struct device *dev, struct kobj_uevent_env *env)
>>   #endif
>>   
>>   #ifdef CONFIG_PARAVIRT
>> +static ssize_t store_paravirt_cpus(struct device *dev,
>> +				   struct device_attribute *attr,
>> +				   const char *buf, size_t count)
>> +{
>> +	cpumask_var_t temp_mask;
>> +	int retval = 0;
>> +
>> +	if (!alloc_cpumask_var(&temp_mask, GFP_KERNEL))
>> +		return -ENOMEM;
>> +
>> +	retval = cpulist_parse(buf, temp_mask);
>> +	if (retval)
>> +		goto free_mask;
>> +
>> +	/* ALL cpus can't be marked as paravirt */
>> +	if (cpumask_equal(temp_mask, cpu_online_mask)) {
>> +		retval = -EINVAL;
>> +		goto free_mask;
>> +	}
>> +	if (cpumask_weight(temp_mask) > num_online_cpus()) {
>> +		retval = -EINVAL;
>> +		goto free_mask;
>> +	}
>> +
>> +	/* No more paravirt cpus */
>> +	if (cpumask_empty(temp_mask)) {
>> +		cpumask_copy((struct cpumask *)&__cpu_paravirt_mask, temp_mask);
>> +	} else {
>> +		cpumask_copy((struct cpumask *)&__cpu_paravirt_mask, temp_mask);
>> +
>> +		/* Enable tick on nohz_full cpu */
>> +		int cpu;
>> +		for_each_cpu(cpu, temp_mask) {
>> +			if (tick_nohz_full_cpu(cpu))
>> +				tick_nohz_dep_set_cpu(cpu, TICK_DEP_BIT_SCHED);
>> +		}
>> +	}
>> +
>> +	retval = count;
>> +
>> +free_mask:
>> +	free_cpumask_var(temp_mask);
>> +	return retval;
>> +}
>> +
>>   static ssize_t print_paravirt_cpus(struct device *dev,
>>   				   struct device_attribute *attr, char *buf)
>>   {
>>   	return sysfs_emit(buf, "%*pbl\n", cpumask_pr_args(cpu_paravirt_mask));
>>   }
>> -static DEVICE_ATTR(paravirt, 0444, print_paravirt_cpus, NULL);
>> +static DEVICE_ATTR(paravirt, 0644, print_paravirt_cpus, store_paravirt_cpus);
> 
> DEVICE_ATTR_RW()?

ok.

> 
> And where is the documentation update for this sysfs file change?
> 

[RFC PATCH v4 11/17] has the documentation of this sysfs file.
https://lore.kernel.org/all/20251119062100.1112520-12-sshegde@linux.ibm.com/

>>   #endif
>>   
>>   const struct bus_type cpu_subsys = {
>> @@ -675,7 +720,6 @@ static void __init cpu_register_vulnerabilities(void)
>>   		put_device(dev);
>>   	}
>>   }
>> -
>>   #else
> 
> Why is this change needed?
> 
> thanks,
> 
> greg k-h

This is a helper patch. This helps to verify functionality of any combination
of CPUs being marked as paravirt which helped me to test some corner cases.

This is also helpful until the arch specific hint becomes better.

This is also useful for other archs which haven't implemented archs specific handling of
steal time, but want to play around with series for their usecase (ex: S390)

Once arch specific hint becomes better, we could decide to remove it or keep in more appropriate
place. It really is debugfs for infra which says I don't want to use these CPUs for now.