Re: [RFC PATCH v2] ptr_ring: linked list fallback

[Jason Wang]

On 2018å02æ28æ 23:43, Michael S. Tsirkin wrote:
> On Wed, Feb 28, 2018 at 10:20:33PM +0800, Jason Wang wrote:
> > On 2018å02æ28æ 22:01, Michael S. Tsirkin wrote:
> > > On Wed, Feb 28, 2018 at 02:28:21PM +0800, Jason Wang wrote:
> > > > On 2018å02æ28æ 12:09, Michael S. Tsirkin wrote:
> > > > > > > Or we can add plist to a union:
> > > > > > >
> > > > > > >
> > > > > > > struct sk_buff {
> > > > > > >             union {
> > > > > > >                     struct {
> > > > > > >                             /* These two members must be first. */
> > > > > > >                             struct sk_buff          *next;
> > > > > > >                             struct sk_buff          *prev;
> > > > > > >                             union {
> > > > > > >                                     struct net_device       *dev;
> > > > > > >                                     /* Some protocols might use this space to store information,
> > > > > > >                                      * while device pointer would be NULL.
> > > > > > >                                      * UDP receive path is one user.
> > > > > > >                                      */
> > > > > > >                                     unsigned long           dev_scratch;
> > > > > > >                             };
> > > > > > >                     };
> > > > > > >                     struct rb_node  rbnode; /* used in netem & tcp stack */
> > > > > > > +		struct plist plist; /* For use with ptr_ring */
> > > > > > >             };
> > > > > > This look ok.
> > > > > >
> > > > > > > > For XDP, we need to embed plist in struct xdp_buff too,
> > > > > > > Right - that's pretty straightforward, isn't it?
> > > > > > Yes, it's not clear to me this is really needed for XDP consider the lock
> > > > > > contention it brings.
> > > > > >
> > > > > > Thanks
> > > > > The contention is only when the ring overflows into the list though.
> > > > Right, but there's usually a mismatch of speed between producer and
> > > > consumer. In case of a fast producer, we may get this contention very
> > > > frequently.
> > > >
> > > > Thanks
> > > This is not true in my experiments.  In my experiments, ring size of 4k
> > > bytes is enough to see packet drops in single %s of cases.
> > >
> > > To you have workloads where rings are full most of the time?
> > E.g using xdp_redirect to redirect packets from ixgbe to tap. In my test,
> > ixgeb can produce ~8Mpps. But vhost can only consume ~3.5Mpps.
> Then you are better off just using a small ring and dropping
> packets early, right?

Yes, so I believe we won't use this for XDP.

Thanks

> > > One other nice side effect of this patch is that instead of dropping
> > > packets quickly it slows down producer to match consumer speeds.
> > In some case, producer may not want to be slowed down, e.g in devmap which
> > can redirect packets into several different interfaces.
> > > IOW, it can go either way in theory, we will need to test and see the effect.
> > Yes.
> >
> > Thanks