[v3,2/4] mbuf: add packet type for UDP-ESP tunnel packets

  Adding new mbuf packet type for UDP encapsulated
ESP packets.

Signed-off-by: Tejasree Kondoj <ktejasree@marvell.com>
---
 doc/guides/rel_notes/release_21_05.rst |  5 +++++
 lib/librte_mbuf/rte_mbuf_ptype.h       | 21 +++++++++++++++++++++
 2 files changed, 26 insertions(+)
  

Hi Olivier, Andrew

Could you please review the patch?

Thanks
Tejasree

> -----Original Message-----
> From: Tejasree Kondoj <ktejasree@marvell.com>
> Sent: Thursday, April 8, 2021 1:47 PM
> To: Akhil Goyal <gakhil@marvell.com>; Radu Nicolau
> <radu.nicolau@intel.com>; Konstantin Ananyev
> <konstantin.ananyev@intel.com>
> Cc: Tejasree Kondoj <ktejasree@marvell.com>; Anoob Joseph
> <anoobj@marvell.com>; Ankur Dwivedi <adwivedi@marvell.com>; Jerin
> Jacob Kollanukkaran <jerinj@marvell.com>; dev@dpdk.org
> Subject: [PATCH v3 2/4] mbuf: add packet type for UDP-ESP tunnel packets
> 
> Adding new mbuf packet type for UDP encapsulated ESP packets.
> 
> Signed-off-by: Tejasree Kondoj <ktejasree@marvell.com>
> ---
>  doc/guides/rel_notes/release_21_05.rst |  5 +++++
>  lib/librte_mbuf/rte_mbuf_ptype.h       | 21 +++++++++++++++++++++
>  2 files changed, 26 insertions(+)
> 
> diff --git a/doc/guides/rel_notes/release_21_05.rst
> b/doc/guides/rel_notes/release_21_05.rst
> index 5565c7637c..c9e9e2ec22 100644
> --- a/doc/guides/rel_notes/release_21_05.rst
> +++ b/doc/guides/rel_notes/release_21_05.rst
> @@ -55,6 +55,11 @@ New Features
>       Also, make sure to start the actual text at the margin.
>       =======================================================
> 
> +* **Added new packet type for UDP-ESP packets in mbuf.**
> +
> +  Added new packet type ``RTE_PTYPE_TUNNEL_ESP_IN_UDP`` which can be
> + used to identify UDP encapsulated ESP packets.
> +
>  * **Enhanced ethdev representor syntax.**
> 
>    * Introduced representor type of VF, SF and PF.
> diff --git a/lib/librte_mbuf/rte_mbuf_ptype.h
> b/lib/librte_mbuf/rte_mbuf_ptype.h
> index 17a2dd3576..bf92ce0c1a 100644
> --- a/lib/librte_mbuf/rte_mbuf_ptype.h
> +++ b/lib/librte_mbuf/rte_mbuf_ptype.h
> @@ -491,6 +491,27 @@ extern "C" {
>   * | 'destination port'=6635>
>   */
>  #define RTE_PTYPE_TUNNEL_MPLS_IN_UDP      0x0000d000
> +/**
> + * ESP-in-UDP tunneling packet type (RFC 3948).
> + *
> + * Packet format:
> + * <'ether type'=0x0800
> + * | 'version'=4, 'protocol'=17
> + * | 'destination port'=4500>
> + * or,
> + * <'ether type'=0x86DD
> + * | 'version'=6, 'next header'=17
> + * | 'destination port'=4500>
> + * or,
> + * <'ether type'=0x0800
> + * | 'version'=4, 'protocol'=17
> + * | 'source port'=4500>
> + * or,
> + * <'ether type'=0x86DD
> + * | 'version'=6, 'next header'=17
> + * | 'source port'=4500>
> + */
> +#define RTE_PTYPE_TUNNEL_ESP_IN_UDP       0x0000e000
>  /**
>   * Mask of tunneling packet types.
>   */
> --
> 2.27.0
  

On Thu, Apr 08, 2021 at 01:47:18PM +0530, Tejasree Kondoj wrote:
> Adding new mbuf packet type for UDP encapsulated
> ESP packets.
> 
> Signed-off-by: Tejasree Kondoj <ktejasree@marvell.com>
> ---
>  doc/guides/rel_notes/release_21_05.rst |  5 +++++
>  lib/librte_mbuf/rte_mbuf_ptype.h       | 21 +++++++++++++++++++++
>  2 files changed, 26 insertions(+)
> 
> diff --git a/doc/guides/rel_notes/release_21_05.rst b/doc/guides/rel_notes/release_21_05.rst
> index 5565c7637c..c9e9e2ec22 100644
> --- a/doc/guides/rel_notes/release_21_05.rst
> +++ b/doc/guides/rel_notes/release_21_05.rst
> @@ -55,6 +55,11 @@ New Features
>       Also, make sure to start the actual text at the margin.
>       =======================================================
>  
> +* **Added new packet type for UDP-ESP packets in mbuf.**
> +
> +  Added new packet type ``RTE_PTYPE_TUNNEL_ESP_IN_UDP`` which can be
> +  used to identify UDP encapsulated ESP packets.
> +
>  * **Enhanced ethdev representor syntax.**
>  
>    * Introduced representor type of VF, SF and PF.
> diff --git a/lib/librte_mbuf/rte_mbuf_ptype.h b/lib/librte_mbuf/rte_mbuf_ptype.h
> index 17a2dd3576..bf92ce0c1a 100644
> --- a/lib/librte_mbuf/rte_mbuf_ptype.h
> +++ b/lib/librte_mbuf/rte_mbuf_ptype.h
> @@ -491,6 +491,27 @@ extern "C" {
>   * | 'destination port'=6635>
>   */
>  #define RTE_PTYPE_TUNNEL_MPLS_IN_UDP      0x0000d000
> +/**
> + * ESP-in-UDP tunneling packet type (RFC 3948).
> + *
> + * Packet format:
> + * <'ether type'=0x0800
> + * | 'version'=4, 'protocol'=17
> + * | 'destination port'=4500>
> + * or,
> + * <'ether type'=0x86DD
> + * | 'version'=6, 'next header'=17
> + * | 'destination port'=4500>
> + * or,
> + * <'ether type'=0x0800
> + * | 'version'=4, 'protocol'=17
> + * | 'source port'=4500>
> + * or,
> + * <'ether type'=0x86DD
> + * | 'version'=6, 'next header'=17
> + * | 'source port'=4500>
> + */
> +#define RTE_PTYPE_TUNNEL_ESP_IN_UDP       0x0000e000
>  /**
>   * Mask of tunneling packet types.
>   */

We arrive at the end of the values in packet type tunnel types,
and there is another pending patch that needs another tunnel type.

As there is already a RTE_PTYPE_TUNNEL_ESP, what would you think about
trying to reuse it, and differentiate IP/ESP from IP/UDP/ESP by using
the L4 layer type (unknown vs udp)? Or maybe add RTE_PTYPE_L4_NONE.

It is sensible, because it can be considered as an API change for
current users of RTE_PTYPE_TUNNEL_ESP. I don't really know how this
type is used by applications.

I think it is time to start thinking about how the packet_type
mbuf API can evolve to solve this issue.

By the way, the update of *rte_get_ptype_tunnel_name() is missing.
  

Hi Olivier,
> On Thu, Apr 08, 2021 at 01:47:18PM +0530, Tejasree Kondoj wrote:
> > Adding new mbuf packet type for UDP encapsulated
> > ESP packets.
> >
> > Signed-off-by: Tejasree Kondoj <ktejasree@marvell.com>
> > ---
> >  doc/guides/rel_notes/release_21_05.rst |  5 +++++
> >  lib/librte_mbuf/rte_mbuf_ptype.h       | 21 +++++++++++++++++++++
> >  2 files changed, 26 insertions(+)
> >
> > diff --git a/doc/guides/rel_notes/release_21_05.rst
> b/doc/guides/rel_notes/release_21_05.rst
> > index 5565c7637c..c9e9e2ec22 100644
> > --- a/doc/guides/rel_notes/release_21_05.rst
> > +++ b/doc/guides/rel_notes/release_21_05.rst
> > @@ -55,6 +55,11 @@ New Features
> >       Also, make sure to start the actual text at the margin.
> >       =======================================================
> >
> > +* **Added new packet type for UDP-ESP packets in mbuf.**
> > +
> > +  Added new packet type ``RTE_PTYPE_TUNNEL_ESP_IN_UDP`` which can
> be
> > +  used to identify UDP encapsulated ESP packets.
> > +
> >  * **Enhanced ethdev representor syntax.**
> >
> >    * Introduced representor type of VF, SF and PF.
> > diff --git a/lib/librte_mbuf/rte_mbuf_ptype.h
> b/lib/librte_mbuf/rte_mbuf_ptype.h
> > index 17a2dd3576..bf92ce0c1a 100644
> > --- a/lib/librte_mbuf/rte_mbuf_ptype.h
> > +++ b/lib/librte_mbuf/rte_mbuf_ptype.h
> > @@ -491,6 +491,27 @@ extern "C" {
> >   * | 'destination port'=6635>
> >   */
> >  #define RTE_PTYPE_TUNNEL_MPLS_IN_UDP      0x0000d000
> > +/**
> > + * ESP-in-UDP tunneling packet type (RFC 3948).
> > + *
> > + * Packet format:
> > + * <'ether type'=0x0800
> > + * | 'version'=4, 'protocol'=17
> > + * | 'destination port'=4500>
> > + * or,
> > + * <'ether type'=0x86DD
> > + * | 'version'=6, 'next header'=17
> > + * | 'destination port'=4500>
> > + * or,
> > + * <'ether type'=0x0800
> > + * | 'version'=4, 'protocol'=17
> > + * | 'source port'=4500>
> > + * or,
> > + * <'ether type'=0x86DD
> > + * | 'version'=6, 'next header'=17
> > + * | 'source port'=4500>
> > + */
> > +#define RTE_PTYPE_TUNNEL_ESP_IN_UDP       0x0000e000
> >  /**
> >   * Mask of tunneling packet types.
> >   */
> 
> We arrive at the end of the values in packet type tunnel types,
> and there is another pending patch that needs another tunnel type.
> 
> As there is already a RTE_PTYPE_TUNNEL_ESP, what would you think about
> trying to reuse it, and differentiate IP/ESP from IP/UDP/ESP by using
> the L4 layer type (unknown vs udp)? Or maybe add RTE_PTYPE_L4_NONE.
> 
> It is sensible, because it can be considered as an API change for
> current users of RTE_PTYPE_TUNNEL_ESP. I don't really know how this
> type is used by applications.

It is OK to use combination of these two but with an assumption
that a normal - IP-UDP packet when encrypted will be an IP-ESP packet
And L4 types are reset from the mbuf->packet_type by the driver.
@Konstantin Ananyev: Are you OK with this assumption?

And, if we choose this path, then also we may need a macro in this file,
So that application doesn't have to combine that explicitly for a standard use case.
#define RTE_PTYPE_TUNNEL_ESP_IN_UDP       RTE_PTYPE_TUNNEL_ESP | RTE_PTYPE_L4_UDP

Will this be fine?

> 
> I think it is time to start thinking about how the packet_type
> mbuf API can evolve to solve this issue.
> 
> By the way, the update of *rte_get_ptype_tunnel_name() is missing.
  

Hi Olivier/ Konstantin,
> > On Thu, Apr 08, 2021 at 01:47:18PM +0530, Tejasree Kondoj wrote:
> > > Adding new mbuf packet type for UDP encapsulated
> > > ESP packets.
> > >
> > > Signed-off-by: Tejasree Kondoj <ktejasree@marvell.com>
> > > ---
> > >  doc/guides/rel_notes/release_21_05.rst |  5 +++++
> > >  lib/librte_mbuf/rte_mbuf_ptype.h       | 21 +++++++++++++++++++++
> > >  2 files changed, 26 insertions(+)
> > >
> > > diff --git a/doc/guides/rel_notes/release_21_05.rst
> > b/doc/guides/rel_notes/release_21_05.rst
> > > index 5565c7637c..c9e9e2ec22 100644
> > > --- a/doc/guides/rel_notes/release_21_05.rst
> > > +++ b/doc/guides/rel_notes/release_21_05.rst
> > > @@ -55,6 +55,11 @@ New Features
> > >       Also, make sure to start the actual text at the margin.
> > >       =======================================================
> > >
> > > +* **Added new packet type for UDP-ESP packets in mbuf.**
> > > +
> > > +  Added new packet type ``RTE_PTYPE_TUNNEL_ESP_IN_UDP`` which can
> > be
> > > +  used to identify UDP encapsulated ESP packets.
> > > +
> > >  * **Enhanced ethdev representor syntax.**
> > >
> > >    * Introduced representor type of VF, SF and PF.
> > > diff --git a/lib/librte_mbuf/rte_mbuf_ptype.h
> > b/lib/librte_mbuf/rte_mbuf_ptype.h
> > > index 17a2dd3576..bf92ce0c1a 100644
> > > --- a/lib/librte_mbuf/rte_mbuf_ptype.h
> > > +++ b/lib/librte_mbuf/rte_mbuf_ptype.h
> > > @@ -491,6 +491,27 @@ extern "C" {
> > >   * | 'destination port'=6635>
> > >   */
> > >  #define RTE_PTYPE_TUNNEL_MPLS_IN_UDP      0x0000d000
> > > +/**
> > > + * ESP-in-UDP tunneling packet type (RFC 3948).
> > > + *
> > > + * Packet format:
> > > + * <'ether type'=0x0800
> > > + * | 'version'=4, 'protocol'=17
> > > + * | 'destination port'=4500>
> > > + * or,
> > > + * <'ether type'=0x86DD
> > > + * | 'version'=6, 'next header'=17
> > > + * | 'destination port'=4500>
> > > + * or,
> > > + * <'ether type'=0x0800
> > > + * | 'version'=4, 'protocol'=17
> > > + * | 'source port'=4500>
> > > + * or,
> > > + * <'ether type'=0x86DD
> > > + * | 'version'=6, 'next header'=17
> > > + * | 'source port'=4500>
> > > + */
> > > +#define RTE_PTYPE_TUNNEL_ESP_IN_UDP       0x0000e000
> > >  /**
> > >   * Mask of tunneling packet types.
> > >   */
> >
> > We arrive at the end of the values in packet type tunnel types,
> > and there is another pending patch that needs another tunnel type.
> >
> > As there is already a RTE_PTYPE_TUNNEL_ESP, what would you think about
> > trying to reuse it, and differentiate IP/ESP from IP/UDP/ESP by using
> > the L4 layer type (unknown vs udp)? Or maybe add RTE_PTYPE_L4_NONE.
> >
> > It is sensible, because it can be considered as an API change for
> > current users of RTE_PTYPE_TUNNEL_ESP. I don't really know how this
> > type is used by applications.
> 
> It is OK to use combination of these two but with an assumption
> that a normal - IP-UDP packet when encrypted will be an IP-ESP packet
> And L4 types are reset from the mbuf->packet_type by the driver.
> @Konstantin Ananyev: Are you OK with this assumption?
> 
> And, if we choose this path, then also we may need a macro in this file,
> So that application doesn't have to combine that explicitly for a standard use
> case.
> #define RTE_PTYPE_TUNNEL_ESP_IN_UDP       RTE_PTYPE_TUNNEL_ESP |
> RTE_PTYPE_L4_UDP
> 
> Will this be fine?
> 
Can we proceed with this approach?

Regards,
Akhil

> >
> > I think it is time to start thinking about how the packet_type
> > mbuf API can evolve to solve this issue.
> >
> > By the way, the update of *rte_get_ptype_tunnel_name() is missing.
  

Hi Akhil,

> 
>  Hi Olivier/ Konstantin,
> > > On Thu, Apr 08, 2021 at 01:47:18PM +0530, Tejasree Kondoj wrote:
> > > > Adding new mbuf packet type for UDP encapsulated
> > > > ESP packets.
> > > >
> > > > Signed-off-by: Tejasree Kondoj <ktejasree@marvell.com>
> > > > ---
> > > >  doc/guides/rel_notes/release_21_05.rst |  5 +++++
> > > >  lib/librte_mbuf/rte_mbuf_ptype.h       | 21 +++++++++++++++++++++
> > > >  2 files changed, 26 insertions(+)
> > > >
> > > > diff --git a/doc/guides/rel_notes/release_21_05.rst
> > > b/doc/guides/rel_notes/release_21_05.rst
> > > > index 5565c7637c..c9e9e2ec22 100644
> > > > --- a/doc/guides/rel_notes/release_21_05.rst
> > > > +++ b/doc/guides/rel_notes/release_21_05.rst
> > > > @@ -55,6 +55,11 @@ New Features
> > > >       Also, make sure to start the actual text at the margin.
> > > >       =======================================================
> > > >
> > > > +* **Added new packet type for UDP-ESP packets in mbuf.**
> > > > +
> > > > +  Added new packet type ``RTE_PTYPE_TUNNEL_ESP_IN_UDP`` which can
> > > be
> > > > +  used to identify UDP encapsulated ESP packets.
> > > > +
> > > >  * **Enhanced ethdev representor syntax.**
> > > >
> > > >    * Introduced representor type of VF, SF and PF.
> > > > diff --git a/lib/librte_mbuf/rte_mbuf_ptype.h
> > > b/lib/librte_mbuf/rte_mbuf_ptype.h
> > > > index 17a2dd3576..bf92ce0c1a 100644
> > > > --- a/lib/librte_mbuf/rte_mbuf_ptype.h
> > > > +++ b/lib/librte_mbuf/rte_mbuf_ptype.h
> > > > @@ -491,6 +491,27 @@ extern "C" {
> > > >   * | 'destination port'=6635>
> > > >   */
> > > >  #define RTE_PTYPE_TUNNEL_MPLS_IN_UDP      0x0000d000
> > > > +/**
> > > > + * ESP-in-UDP tunneling packet type (RFC 3948).
> > > > + *
> > > > + * Packet format:
> > > > + * <'ether type'=0x0800
> > > > + * | 'version'=4, 'protocol'=17
> > > > + * | 'destination port'=4500>
> > > > + * or,
> > > > + * <'ether type'=0x86DD
> > > > + * | 'version'=6, 'next header'=17
> > > > + * | 'destination port'=4500>
> > > > + * or,
> > > > + * <'ether type'=0x0800
> > > > + * | 'version'=4, 'protocol'=17
> > > > + * | 'source port'=4500>
> > > > + * or,
> > > > + * <'ether type'=0x86DD
> > > > + * | 'version'=6, 'next header'=17
> > > > + * | 'source port'=4500>
> > > > + */
> > > > +#define RTE_PTYPE_TUNNEL_ESP_IN_UDP       0x0000e000
> > > >  /**
> > > >   * Mask of tunneling packet types.
> > > >   */
> > >
> > > We arrive at the end of the values in packet type tunnel types,
> > > and there is another pending patch that needs another tunnel type.
> > >
> > > As there is already a RTE_PTYPE_TUNNEL_ESP, what would you think about
> > > trying to reuse it, and differentiate IP/ESP from IP/UDP/ESP by using
> > > the L4 layer type (unknown vs udp)? Or maybe add RTE_PTYPE_L4_NONE.
> > >
> > > It is sensible, because it can be considered as an API change for
> > > current users of RTE_PTYPE_TUNNEL_ESP. I don't really know how this
> > > type is used by applications.
> >
> > It is OK to use combination of these two but with an assumption
> > that a normal - IP-UDP packet when encrypted will be an IP-ESP packet
> > And L4 types are reset from the mbuf->packet_type by the driver.
> > @Konstantin Ananyev: Are you OK with this assumption?
> >
> > And, if we choose this path, then also we may need a macro in this file,
> > So that application doesn't have to combine that explicitly for a standard use
> > case.
> > #define RTE_PTYPE_TUNNEL_ESP_IN_UDP       RTE_PTYPE_TUNNEL_ESP |
> > RTE_PTYPE_L4_UDP
> >
> > Will this be fine?
> >
> Can we proceed with this approach?

I think we can safely use such combination inside ipsec-secgw app.
About making it a new generic type - I am not so sure. 
As Olivier already pointed out - it looks like an API/behaviour breakage to me. 

> Regards,
> Akhil
> 
> > >
> > > I think it is time to start thinking about how the packet_type
> > > mbuf API can evolve to solve this issue.

+1
Might be it needs to be reworked completely.

> > >
> > > By the way, the update of *rte_get_ptype_tunnel_name() is missing.