Re: [PATCH v2 7/7] KVM: x86/mmu: Lazily allocate memslot rmaps

[Paolo Bonzini]

On 29/04/21 23:18, Ben Gardon wrote:
> +	/*
> +	 * If set, the rmap should be allocated for any newly created or
> +	 * modified memslots. If allocating rmaps lazily, this may be set
> +	 * before the rmaps are allocated for existing memslots, but
> +	 * shadow_mmu_active will not be set until after the rmaps are fully
> +	 * allocated.
> +	 */
> +	bool alloc_memslot_rmaps;

Let's remove the whole sentence starting with "If allocating rmaps 
lazily".  The part about shadow_mmu_active should go there, while the 
rest is pointless as long as we just say that this flag will be accessed 
only under slots_arch_lock.

(Regarding shadow_mmu_active, I think I know what Sean will be 
suggesting because I had a similar thought and decided it introduced 
extra unnecessary complication... but maybe not, so let's see what he says).

Paolo