Re: [RFC PATCH] mm, oom: distinguish blockable mode for mmu notifiers

[Paolo Bonzini]

On 25/06/2018 10:45, Michal Hocko wrote:
> On Mon 25-06-18 10:10:18, Paolo Bonzini wrote:
>> On 25/06/2018 09:57, Michal Hocko wrote:
>>> On Sun 24-06-18 10:11:21, Paolo Bonzini wrote:
>>>> On 22/06/2018 17:02, Michal Hocko wrote:
>>>>> @@ -7215,6 +7216,8 @@ void kvm_arch_mmu_notifier_invalidate_range(struct kvm *kvm,
>>>>>  	apic_address = gfn_to_hva(kvm, APIC_DEFAULT_PHYS_BASE >> PAGE_SHIFT);
>>>>>  	if (start <= apic_address && apic_address < end)
>>>>>  		kvm_make_all_cpus_request(kvm, KVM_REQ_APIC_PAGE_RELOAD);
>>>>> +
>>>>> +	return 0;
>>>>
>>>> This is wrong, gfn_to_hva can sleep.
>>>
>>> Hmm, I have tried to crawl the call chain and haven't found any
>>> sleepable locks taken. Maybe I am just missing something.
>>> __kvm_memslots has a complex locking assert. I do not see we would take
>>> slots_lock anywhere from the notifier call path. IIUC that means that
>>> users_count has to be zero at that time. I have no idea how that is
>>> guaranteed.
>>
>> Nevermind, ENOCOFFEE.  This is gfn_to_hva, not gfn_to_pfn.  It only
>> needs SRCU.
> 
> OK, so just the make sure I follow, the change above is correct?

Yes.  It's only gfn_to_pfn that calls get_user_pages and therefore can
sleep.

Thanks,

Paolo