Re: [RFC 13/14] KVM: x86/MMU: Wrap uses of kvm_handle_gfn_range in mmu.c

[Sean Christopherson]

 On Wed, Dec 21, 2022, Ben Gardon wrote:
> @@ -978,9 +978,13 @@ static void slot_rmap_walk_next(struct slot_rmap_walk_iterator *iterator)
>  	     slot_rmap_walk_okay(_iter_);				\
>  	     slot_rmap_walk_next(_iter_))
>  
> -__always_inline bool kvm_handle_gfn_range(struct kvm *kvm,
> -					  struct kvm_gfn_range *range,
> -					  rmap_handler_t handler)
> +typedef bool (*rmap_handler_t)(struct kvm *kvm, struct kvm_rmap_head *rmap_head,
> +			       struct kvm_memory_slot *slot, gfn_t gfn,
> +			       int level, pte_t pte);
> +
> +static __always_inline bool
> +kvm_handle_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range,

Don't split function returns/attributes from the function declaration.  I don't
think the rule ended up getting officially documented and enforced, but Linus was
unequivocal when it came up[*], and I happen to agree with him :-)

Actually, since I'm guessing you got the idea from existing code, can you fold
in the attached patches to purge the existing cases in mmu.c before those uglies
get moved around?  Assuming you don't dislike the proposed rename, that is.

[*] https://lore.kernel.org/mm-commits/CAHk-=wjS-Jg7sGMwUPpDsjv392nDOOs0CtUtVkp=S6Q7JzFJRw@xxxxxxxxxxxxxx