> On 7 February 2017 at 12:22, Stokes, Ian <[email protected]> wrote: > >> On 6 February 2017 at 09:20, Ian Stokes <[email protected]> wrote: > >> > Remove the experimental warning tag in documentation regarding OVS > >> > deployed with DPDK. > >> > > >> > Signed-off-by: Ian Stokes <[email protected]> > >> > >> <snip> > >> > >> > diff --git a/README.rst b/README.rst index f5cdaa5..afdd9bc 100644 > >> > --- a/README.rst > >> > +++ b/README.rst > >> > @@ -38,9 +38,11 @@ following features: > >> > > >> > The included Linux kernel module supports Linux 3.10 and up. > >> > > >> > -Open vSwitch can also operate, at a cost in performance, entirely > >> > in userspace, -without assistance from a kernel module. This > >> > userspace implementation should -be easier to port than the kernel- > based switch. > >> It is considered experimental. > >> > +Open vSwitch can also operate entirely in userspace without > >> > +assistance from a kernel module. This userspace implementation > >> > +should be easier to port than the kernel-based switch. OVS in > >> > +userspace > >> can access Linux or DPDK devices. > >> > +Note Open vSwitch with Linux devices in userspace is considered > >> > +experimental and comes with a cost in performance. > >> > >> I wonder if we can clarify this last sentence, it appears to be > >> causing some confusion about the status of the kernel datapath. > > > > Do you mean it causes confunsion in terms of performance of kernel space > vs user space? > > Not related to performance, but whether kernel datapath is considered > experimental or not. > > > If that’s the case we could change it to > > > > +Note Open vSwitch with Linux devices in userspace is considered > > +experimental and comes with a cost in performance when compared to > kernel space. > > I think the problem is "Open vSwitch with Linux devices in userspace". > This doesn't precisely define usage of the userspace datapath. > > Perhaps this? > > Note Open vSwitch with userspace datapath and non-DPDK devices is > considered experimental and comes with a cost in performance.
Agreed, that seems clearer. I'll submit a patch with the change. Ian _______________________________________________ dev mailing list [email protected] https://mail.openvswitch.org/mailman/listinfo/ovs-dev
