[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [PATCH v3] virtio_net: support split header
On Tue, 14 Jun 2022 16:51:58 +0800, Jason Wang <jasowang@redhat.com> wrote:
> On Mon, Jun 6, 2022 at 8:11 PM Xuan Zhuo <xuanzhuo@linux.alibaba.com> wrote:
> >
> > The purpose of this feature is to split the header and the payload of
> > the packet.
> >
> > | receive buffer |
> > | 0th descriptor | 1th descriptor |
> > | virtnet hdr | mac | ip hdr | tcp hdr|<-- hold -->| payload |
> >
> > We can use a buffer plus a separate page when allocating the receive
> > buffer. In this way, we can ensure that all payloads can be
> > independently in a page, which is very beneficial for the zerocopy
> > implemented by the upper layer.
> >
> > Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
> > ---
> >
> > 1. Fix some syntax issues
> > 2. Fix some terminology issues
> > 3. It is not unified with ip alignment, so ip alignment is not included
> > 4. Make it clear that the device must support four types, in the case of
> > successful negotiation.
> >
> > conformance.tex | 2 ++
> > content.tex | 82 +++++++++++++++++++++++++++++++++++++++++++++++++
> > 2 files changed, 84 insertions(+)
> >
> > diff --git a/conformance.tex b/conformance.tex
> > index 663e7c3..6f561fb 100644
> > --- a/conformance.tex
> > +++ b/conformance.tex
> > @@ -149,6 +149,7 @@ \section{Conformance Targets}\label{sec:Conformance / Conformance Targets}
> > \item \ref{drivernormative:Device Types / Network Device / Device Operation / Control Virtqueue / Automatic receive steering in multiqueue mode}
> > \item \ref{drivernormative:Device Types / Network Device / Device Operation / Control Virtqueue / Offloads State Configuration / Setting Offloads State}
> > \item \ref{drivernormative:Device Types / Network Device / Device Operation / Control Virtqueue / Receive-side scaling (RSS) }
> > +\item \ref{drivernormative:Device Types / Network Device / Device Operation / Control Virtqueue / Split Header}
> > \end{itemize}
> >
> > \conformance{\subsection}{Block Driver Conformance}\label{sec:Conformance / Driver Conformance / Block Driver Conformance}
> > @@ -411,6 +412,7 @@ \section{Conformance Targets}\label{sec:Conformance / Conformance Targets}
> > \item \ref{devicenormative:Device Types / Network Device / Device Operation / Control Virtqueue / Gratuitous Packet Sending}
> > \item \ref{devicenormative:Device Types / Network Device / Device Operation / Control Virtqueue / Automatic receive steering in multiqueue mode}
> > \item \ref{devicenormative:Device Types / Network Device / Device Operation / Control Virtqueue / Receive-side scaling (RSS) / RSS processing}
> > +\item \ref{devicenormative:Device Types / Network Device / Device Operation / Control Virtqueue / Split Header}
> > \end{itemize}
> >
> > \conformance{\subsection}{Block Device Conformance}\label{sec:Conformance / Device Conformance / Block Device Conformance}
> > diff --git a/content.tex b/content.tex
> > index 7508dd1..a5a29d2 100644
> > --- a/content.tex
> > +++ b/content.tex
> > @@ -3084,6 +3084,9 @@ \subsection{Feature bits}\label{sec:Device Types / Network Device / Feature bits
> > \item[VIRTIO_NET_F_CTRL_MAC_ADDR(23)] Set MAC address through control
> > channel.
> >
> > +\item[VIRTIO_NET_F_SPLIT_HEADER (55)] Device supports to split the protocol
> > + header and the payload.
> > +
> > \item[VIRTIO_NET_F_HOST_USO (56)] Device can receive USO packets. Unlike UFO
> > (fragmenting the packet) the USO splits large UDP packet
> > to several segments when each of these smaller packets has UDP header.
> > @@ -3131,6 +3134,7 @@ \subsubsection{Feature bit requirements}\label{sec:Device Types / Network Device
> > \item[VIRTIO_NET_F_CTRL_MAC_ADDR] Requires VIRTIO_NET_F_CTRL_VQ.
> > \item[VIRTIO_NET_F_RSC_EXT] Requires VIRTIO_NET_F_HOST_TSO4 or VIRTIO_NET_F_HOST_TSO6.
> > \item[VIRTIO_NET_F_RSS] Requires VIRTIO_NET_F_CTRL_VQ.
> > +\item[VIRTIO_NET_F_SPLIT_HEADER] Requires VIRTIO_NET_F_CTRL_VQ.
> > \end{description}
> >
> > \subsubsection{Legacy Interface: Feature bits}\label{sec:Device Types / Network Device / Feature bits / Legacy Interface: Feature bits}
> > @@ -3362,6 +3366,7 @@ \subsection{Device Operation}\label{sec:Device Types / Network Device / Device O
> > #define VIRTIO_NET_HDR_F_NEEDS_CSUM 1
> > #define VIRTIO_NET_HDR_F_DATA_VALID 2
> > #define VIRTIO_NET_HDR_F_RSC_INFO 4
> > +#define VIRTIO_NET_HDR_F_SPLIT_HEADER 8
> > u8 flags;
> > #define VIRTIO_NET_HDR_GSO_NONE 0
> > #define VIRTIO_NET_HDR_GSO_TCPV4 1
> > @@ -4463,6 +4468,83 @@ \subsubsection{Control Virtqueue}\label{sec:Device Types / Network Device / Devi
> > according to the native endian of the guest rather than
> > (necessarily when not using the legacy interface) little-endian.
> >
> > +\paragraph{Split Header}\label{sec:Device Types / Network Device / Device Operation / Control Virtqueue / Split Header}
> > +
> > +If the VIRTIO_NET_F_SPLIT_HEADER feature is negotiated,
> > +the device supports to split the protocol header and the payload.
> > +The protocol header and the payload will be separated into different
> > +descriptors.
> > +
> > +\subparagraph{Split Header}\label{sec:Device Types / Network Device / Device Operation / Control Virtqueue / Split Header / Setting Split Header}
> > +
> > +To configure the split header, the following layout structure and definitions
> > +are used:
> > +
> > +\begin{lstlisting}
> > +struct virtio_net_split_header_config {
> > +#define VIRTIO_NET_SPLIT_HEADER_TYPE_TCP4 1
> > +#define VIRTIO_NET_SPLIT_HEADER_TYPE_TCP6 2
> > +#define VIRTIO_NET_SPLIT_HEADER_TYPE_UDP4 4
> > +#define VIRTIO_NET_SPLIT_HEADER_TYPE_UDP6 8
> > + le64 type;
> > +};
> > +
> > +#define VIRTIO_NET_CTRL_SPLIT_HEADER 6
> > + #define VIRTIO_NET_CTRL_SPLIT_HEADER_SET 0
> > +\end{lstlisting}
> > +
> > +The class VIRTIO_NET_CTRL_SPLIT_HEADER has one command:
> > +VIRTIO_NET_CTRL_SPLIT_HEADER_SET applies the new split header configuration.
> > +
> > +The driver can enable or disable the split header by setting or clearing
> > +corresponding bits in \field{type}.
> > +
> > +The protocol header contains all headers before level 4 payload.
>
> So if the split header means only for level 4 we need to clarify it in
> the above feature definition.
>
> Or is it worthwhile to leave space to split at level 3? (E.g modern
> NICs can do splitting at various layer, l2/l3 and inner)
Should support splitting at various layer. It's my expression problem.
>
> > For ip
> > +fragmented packets, except for the first fragment, the protocol header only
> > +contains all headers before level 3 payload.
>
> Does it mean the device needs to track the fragment ID and only do
> this for the L4 protocol that is set via
> VIRTIO_NET_CTRL_SPLIT_HEADER_SET?
>
> This looks complicates the design of the device and it doesn't help
> too much since most segmentations are smaller than page size
I agree. So I think it should not be supported for split ip fragmented packets
>
> > +
> > +\devicenormative{\subparagraph}{Setting Split Header}{Device Types / Network Device / Device Operation / Control Virtqueue / Split Header}
> > +
> > +Split header MUST be disabled upon device reset.
> > +
> > +If VIRTIO_NET_F_SPLIT_HEADER is negotiated, the device MUST support
> > +VIRTIO_NET_SPLIT_HEADER_TYPE_TCP4, VIRTIO_NET_SPLIT_HEADER_TYPE_TCP6,
> > +VIRTIO_NET_SPLIT_HEADER_TYPE_UDP4, VIRTIO_NET_SPLIT_HEADER_TYPE_UDP6.
> > +
> > +A device MUST NOT perform split header in the following cases:
> > +\begin{itemize}
> > + \item device does not recognize the protocol of the packet.
> > + \item \field{type} does not include the protocol of the packet.
> > + \item the buffer consists of only one descriptor.
> > + \item the virtio net header and the protocol header exceeds the size
> > + of the first descriptor.
> > + \item If VIRTIO_NET_F_MRG_RXBUF is not negotiated and the size of the
> > + payload exceeds the size of the buffer chain starting from the 2nd
>
> I think it's a "descriptor chain" instead of a buffer chain, since we
> can't use the buffer chain if MRG_RXBUF is not negotiated.
Will fix.
>
> > + descriptor.
> > +\end{itemize}
> > +
> > +If the header is split by the device, the \field{flags} of structure
> > +virtio_net_hdr MUST contains VIRTIO_NET_HDR_F_SPLIT_HEADER. The protocol header
>
> s/contains/contain/
Will fix.
>
> > +MUST be on the buffer specified by the first descriptor, following the virtio
> > +net header. The payload MUST starts from the buffer of the second descriptor.
>
> s/starts/start/, I think we can remove "the buffer of" otherwise it
> may be confusing.
Will fix.
>
> > +The device MUST set \field{hdr_len} of structure virtio_net_hdr to the length of
> > +the protocol header.
> > +
> > +If the header is split by the device, VIRTIO_NET_F_MRG_RXBUF is negotiated,
> > +and the receive packet is spread over multiple buffers, when the device uses a
> > +buffer other than the first, if the buffer consists of multiple descriptors, the
> > +device MUST skip the buffer of the first descriptor,
>
> skip the first descriptor of the buffer actually?
Will fix.
Thanks.
>
> > because the buffer of the
> > +first descriptor is used to carry the protocol header.
> > +
> > +\drivernormative{\subparagraph}{Setting Split Header}{Device Types / Network Device / Device Operation / Control Virtqueue / Split Header}
> > +
> > +If VIRTIO_NET_HDR_F_SPLIT_HEADER bit in \field{flags} is set, the driver MUST
> > +trust \field{hdr_len}.
> > +
> > +If the split header is enabled, the buffers submitted to receiveq by the
> > +driver SHOULD at least be composed of two descriptors. The buffer specified by
> > +the first descriptor SHOULD be able to accommodate the virtio net header and the
> > +protocol header.
>
> Thanks
>
> >
> > \subsubsection{Legacy Interface: Framing Requirements}\label{sec:Device
> > Types / Network Device / Legacy Interface: Framing Requirements}
> > --
> > 2.31.0
> >
>
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]