Re: [RFC net-next 2/9] xdp: add XDP_FLAGS_GRO_DISABLED flag

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Fri, Jun 21, 2024 at 4:17 AM Alexander Lobakin
<aleksander.lobakin@xxxxxxxxx> wrote:
>
> From: Yan Zhai <yan@xxxxxxxxxxxxxx>
> Date: Thu, 20 Jun 2024 15:19:13 -0700
>
> > Allow XDP program to set XDP_FLAGS_GRO_DISABLED flag in xdp_buff and
> > xdp_frame, and disable GRO when building an sk_buff if this flag is set.
> >
> > Signed-off-by: Yan Zhai <yan@xxxxxxxxxxxxxx>
> > ---
> >  include/net/xdp.h | 38 +++++++++++++++++++++++++++++++++++++-
> >  1 file changed, 37 insertions(+), 1 deletion(-)
> >
> > diff --git a/include/net/xdp.h b/include/net/xdp.h
> > index e6770dd40c91..cc3bce8817b0 100644
> > --- a/include/net/xdp.h
> > +++ b/include/net/xdp.h
> > @@ -75,6 +75,7 @@ enum xdp_buff_flags {
> >       XDP_FLAGS_FRAGS_PF_MEMALLOC     = BIT(1), /* xdp paged memory is under
> >                                                  * pressure
> >                                                  */
> > +     XDP_FLAGS_GRO_DISABLED          = BIT(2), /* GRO disabled */
>
> There should be tabs, not spaces.
>
> >  };
> >
> >  struct xdp_buff {
> > @@ -113,12 +114,35 @@ static __always_inline void xdp_buff_set_frag_pfmemalloc(struct xdp_buff *xdp)
> >       xdp->flags |= XDP_FLAGS_FRAGS_PF_MEMALLOC;
> >  }
> >
> > +static __always_inline void xdp_buff_disable_gro(struct xdp_buff *xdp)
> > +{
> > +     xdp->flags |= XDP_FLAGS_GRO_DISABLED;
> > +}
> > +
> > +static __always_inline bool xdp_buff_gro_disabled(struct xdp_buff *xdp)
> > +{
> > +     return !!(xdp->flags & XDP_FLAGS_GRO_DISABLED);
> > +}
> > +
> > +static __always_inline void
> > +xdp_buff_fixup_skb_offloading(struct xdp_buff *xdp, struct sk_buff *skb)
> > +{
> > +     if (xdp_buff_gro_disabled(xdp))
> > +             skb_disable_gro(skb);
> > +}
>
> I don't think this should be named "fixup". "propagate", "update",
> "promote", ...?
>
> Maybe `if` is not needed here?
>
>         skb->gro_disabled = xdp_buff_gro_disabled(xdp)
>
> ?
>
I called it fixup mainly considering current skb fields could be set
incorrectly (or maybe not?) For example when XDP load balances
traffic, it could encap and send from server A to server B, but that
means HW on B may have no good way to calculate the actual flow
hash/csum. The original values could be read from A and sent within
the encapsulation header, so that B can use this value to "fixup" the
things.

But tbh, I am never good at naming, it's too hard for a non-native
speaker sometimes... So I am very open on what to use if majority of
the participants like it :D

The if is still needed since I added a control config to wrap that bit.

best
Yan

> > +
> > +static __always_inline void
> > +xdp_init_buff_minimal(struct xdp_buff *xdp)
> > +{
> > +     xdp->flags = 0;
> > +}
>
> "xdp_buff_clear_flags"?
>
> > +
> >  static __always_inline void
> >  xdp_init_buff(struct xdp_buff *xdp, u32 frame_sz, struct xdp_rxq_info *rxq)
> >  {
> >       xdp->frame_sz = frame_sz;
> >       xdp->rxq = rxq;
> > -     xdp->flags = 0;
> > +     xdp_init_buff_minimal(xdp);
> >  }
> >
> >  static __always_inline void
> > @@ -187,6 +211,18 @@ static __always_inline bool xdp_frame_is_frag_pfmemalloc(struct xdp_frame *frame
> >       return !!(frame->flags & XDP_FLAGS_FRAGS_PF_MEMALLOC);
> >  }
> >
> > +static __always_inline bool xdp_frame_gro_disabled(struct xdp_frame *frame)
> > +{
> > +     return !!(frame->flags & XDP_FLAGS_GRO_DISABLED);
> > +}
> > +
> > +static __always_inline void
> > +xdp_frame_fixup_skb_offloading(struct xdp_frame *frame, struct sk_buff *skb)
> > +{
> > +     if (xdp_frame_gro_disabled(frame))
> > +             skb_disable_gro(skb);
> > +}
>
> (same)
>
> > +
> >  #define XDP_BULK_QUEUE_SIZE  16
> >  struct xdp_frame_bulk {
> >       int count;
>
> Thanks,
> Olek





[Index of Archives]     [Linux Samsung SoC]     [Linux Rockchip SoC]     [Linux Actions SoC]     [Linux for Synopsys ARC Processors]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]


  Powered by Linux