Re: [PATCH V6 6/6] kvm: add fast mmio capabilitiy

[Cornelia Huck <cornelia.huck@de.ibm.com>]

On Tue, 15 Sep 2015 17:07:55 +0200
Paolo Bonzini <pbonzini@redhat.com> wrote:

> On 15/09/2015 08:41, Jason Wang wrote:
> > +With KVM_CAP_FAST_MMIO, a zero length mmio eventfd is allowed for
> > +kernel to ignore the length of guest write and get a possible faster
> > +response. Note the speedup may only work on some specific
> > +architectures and setups. Otherwise, it's as fast as wildcard mmio
> > +eventfd.
> 
> I don't really like tying the capability to MMIO, especially since
> zero length ioeventfd is already accepted for virtio-ccw.

Actually, zero length ioeventfd does not make sense for virtio-ccw; we
just don't check it (although we probably should).

> 
> What about the following?
> 
> diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt
> index 7a3cb48a644d..247944071cc8 100644
> --- a/Documentation/virtual/kvm/api.txt
> +++ b/Documentation/virtual/kvm/api.txt
> @@ -1627,11 +1627,10 @@ to the registered address is equal to datamatch in struct kvm_ioeventfd.
>  For virtio-ccw devices, addr contains the subchannel id and datamatch the
>  virtqueue index.
> 
> -With KVM_CAP_FAST_MMIO, a zero length mmio eventfd is allowed for
> -kernel to ignore the length of guest write and get a possible faster
> -response. Note the speedup may only work on some specific
> -architectures and setups. Otherwise, it's as fast as wildcard mmio
> -eventfd.
> +With KVM_CAP_IOEVENTFD_ANY_LENGTH, a zero length ioeventfd is allowed, and
> +the kernel will ignore the length of guest write and get a faster vmexit.

s/get/may get/ ?

> +The speedup may only apply to specific architectures, but the ioeventfd will
> +work anyway.