[PATCH v13 34/85] KVM: Get writable mapping for __kvm_vcpu_map() only when necessary

[Yan Zhao]

On Thu, Oct 10, 2024 at 11:23:36AM -0700, Sean Christopherson wrote:
> When creating a memory map for read, don't request a writable pfn from the
> primary MMU.  While creating read-only mappings can be theoretically slower,
> as they don't play nice with fast GUP due to the need to break CoW before
> mapping the underlying PFN, practically speaking, creating a mapping isn't
> a super hot path, and getting a writable mapping for reading is weird and
> confusing.
> 
> Tested-by: Alex Bennée <alex.bennee at linaro.org>
> Signed-off-by: Sean Christopherson <seanjc at google.com>
> ---
>  virt/kvm/kvm_main.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
> index 080740f65061..b845e9252633 100644
> --- a/virt/kvm/kvm_main.c
> +++ b/virt/kvm/kvm_main.c
> @@ -3122,7 +3122,7 @@ int __kvm_vcpu_map(struct kvm_vcpu *vcpu, gfn_t gfn, struct kvm_host_map *map,
>  	struct kvm_follow_pfn kfp = {
>  		.slot = gfn_to_memslot(vcpu->kvm, gfn),
>  		.gfn = gfn,
> -		.flags = FOLL_WRITE,
> +		.flags = writable ? FOLL_WRITE : 0,
>  		.refcounted_page = &map->pinned_page,
>  		.pin = true,
>  	};
When writable is false, could we set ".pin = false," ?

Also not sure if ".map_writable = NULL" is missing.