Re: [PATCH] KVM: x86/mmu: Add "never" option to allow sticky disabling of nx_huge_pages

[Robert Hoo <robert.hoo.linux@gmail.com>]

On 6/2/2023 11:09 PM, Sean Christopherson wrote:
> On Fri, Jun 02, 2023, Robert Hoo wrote:
>> On 6/2/2023 8:58 AM, Sean Christopherson wrote:
>>> @@ -6860,15 +6871,29 @@ static int set_nx_huge_pages(const char *val, const struct kernel_param *kp)
>>>    	bool old_val = nx_huge_pages;
>>>    	bool new_val;
>>> +	if (nx_hugepage_mitigation_hard_disabled)
>>> +		return -EPERM;
>>> +
>>>    	/* In "auto" mode deploy workaround only if CPU has the bug. */
>>> -	if (sysfs_streq(val, "off"))
>>> +	if (sysfs_streq(val, "off")) {
>>>    		new_val = 0;
>>> -	else if (sysfs_streq(val, "force"))
>>> +	} else if (sysfs_streq(val, "force")) {
>>>    		new_val = 1;
>>> -	else if (sysfs_streq(val, "auto"))
>>> +	} else if (sysfs_streq(val, "auto")) {
>>>    		new_val = get_nx_auto_mode();
>>> -	else if (kstrtobool(val, &new_val) < 0)
>>> +	} else if (sysfs_streq(val, "never")) {
>>> +		new_val = 0;
>>> +
>>> +		mutex_lock(&kvm_lock);
>>> +		if (!list_empty(&vm_list)) {
>>> +			mutex_unlock(&kvm_lock);
>>> +			return -EBUSY;

Ah, right, thanks for explanation.

Reviewed-by: Robert Hoo <robert.hoo.linux@gmail.com>

>>> +		}
>>> +		nx_hugepage_mitigation_hard_disabled = true;
>>> +		mutex_unlock(&kvm_lock);
>>> +	} else if (kstrtobool(val, &new_val) < 0) {
>>>    		return -EINVAL;
>>> +	}
>>>
>>
>> IIUC, (Initially) "auto_off"/"off" --> create some VM --> turn to "never",
>> the created VMs still have those kthreads, but can never be used, until
>> destroyed with VM.
> 
> Shouldn't be able to happen.  The above rejects "never" if vm_list isn't empty,
> i.e. if there are any VMs, and sets nx_hugepage_mitigation_hard_disabled under
> kvm_lock to ensure it can't race with KVM_CREATE_VM.  I forgot to call this out
> in the changelog though