[dpdk-dev,v2,2/6] ethdev: add GTPC and GTPU items

  This patch adds GTPC and GTPU items to generic rte
flow, and also exposes the following item fields
through the flow command:

- GTPC TEID
- GTPU TEID

Signed-off-by: Beilei Xing <beilei.xing@intel.com>
---
 app/test-pmd/cmdline_flow.c                 | 44 +++++++++++++++++++++++++++++
 app/test-pmd/config.c                       |  2 ++
 doc/guides/prog_guide/rte_flow.rst          | 26 +++++++++++++++++
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |  8 ++++++
 lib/librte_ether/rte_flow.h                 | 44 +++++++++++++++++++++++++++++
 5 files changed, 124 insertions(+)
  

Hi Beilei,

I assume this patch supersedes [1]?

[1] http://dpdk.org/ml/archives/dev/2017-August/073501.html

Thanks for merging testpmd and the API change as a single patch, I still
have a few comments, see below.

(please add "flow API" somewhere in the title by the way)

On Thu, Sep 07, 2017 at 07:20:59PM +0800, Beilei Xing wrote:
> This patch adds GTPC and GTPU items to generic rte
> flow, and also exposes the following item fields
> through the flow command:
> 
> - GTPC TEID
> - GTPU TEID
> 
> Signed-off-by: Beilei Xing <beilei.xing@intel.com>

Won't there be a need to match nonspecific GTP traffic as well (both GTP-C
and GTP-U a once), since they use the same structure?

I'm not familiar with the protocol at all so I wonder if you should maybe
leave the GTP item in addition to those two.

> ---
>  app/test-pmd/cmdline_flow.c                 | 44 +++++++++++++++++++++++++++++
>  app/test-pmd/config.c                       |  2 ++
>  doc/guides/prog_guide/rte_flow.rst          | 26 +++++++++++++++++
>  doc/guides/testpmd_app_ug/testpmd_funcs.rst |  8 ++++++
>  lib/librte_ether/rte_flow.h                 | 44 +++++++++++++++++++++++++++++
>  5 files changed, 124 insertions(+)
> 
> diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c
> index a17a004..72d159c 100644
> --- a/app/test-pmd/cmdline_flow.c
> +++ b/app/test-pmd/cmdline_flow.c
> @@ -171,6 +171,10 @@ enum index {
>  	ITEM_GRE_PROTO,
>  	ITEM_FUZZY,
>  	ITEM_FUZZY_THRESH,
> +	ITEM_GTPC,
> +	ITEM_GTPC_TEID,
> +	ITEM_GTPU,
> +	ITEM_GTPU_TEID,

You could refactor the TEID parameter since they use the same
structure. Might be useful if you add nonspecific GTP:

 ITEM_GTP,
 ITEM_GTP_TEID,
 ITEM_GTPC,
 ITEM_GTPU,

>  
>  	/* Validate/create actions. */
>  	ACTIONS,
> @@ -451,6 +455,8 @@ static const enum index next_item[] = {
>  	ITEM_MPLS,
>  	ITEM_GRE,
>  	ITEM_FUZZY,
> +	ITEM_GTPC,
> +	ITEM_GTPU,
>  	ZERO,
>  };
>  
> @@ -588,6 +594,18 @@ static const enum index item_gre[] = {
>  	ZERO,
>  };
>  
> +static const enum index item_gtpc[] = {
> +	ITEM_GTPC_TEID,
> +	ITEM_NEXT,
> +	ZERO,
> +};
> +
> +static const enum index item_gtpu[] = {
> +	ITEM_GTPU_TEID,
> +	ITEM_NEXT,
> +	ZERO,
> +};

A single array is necessary, item_gtp[].

> +
>  static const enum index next_action[] = {
>  	ACTION_END,
>  	ACTION_VOID,
> @@ -1421,6 +1439,32 @@ static const struct token token_list[] = {
>  		.args = ARGS(ARGS_ENTRY(struct rte_flow_item_fuzzy,
>  					thresh)),
>  	},
> +	[ITEM_GTPC] = {
> +		.name = "gtpc",
> +		.help = "match GTP header",
> +		.priv = PRIV_ITEM(GTPC, sizeof(struct rte_flow_item_gtp)),
> +		.next = NEXT(item_gtpc),
> +		.call = parse_vc,
> +	},
> +	[ITEM_GTPC_TEID] = {
> +		.name = "teid",
> +		.help = "TUNNEL ENDPOINT IDENTIFIER",

Please don't shout, "tunnel endpoint identifier" is fine.

> +		.next = NEXT(item_gtpc, NEXT_ENTRY(UNSIGNED), item_param),
> +		.args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_gtp, teid)),
> +	},
> +	[ITEM_GTPU] = {
> +		.name = "gtpu",
> +		.help = "match GTP header",
> +		.priv = PRIV_ITEM(GTPU, sizeof(struct rte_flow_item_gtp)),
> +		.next = NEXT(item_gtpu),
> +		.call = parse_vc,
> +	},
> +	[ITEM_GTPU_TEID] = {
> +		.name = "teid",
> +		.help = "TUNNEL ENDPOINT IDENTIFIER",

Same comment here, however the a single TEID entry is necessary as
previously described.

> +		.next = NEXT(item_gtpu, NEXT_ENTRY(UNSIGNED), item_param),
> +		.args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_gtp, teid)),
> +	},
>  
>  	/* Validate/create actions. */
>  	[ACTIONS] = {
> diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
> index 3ae3e1c..be4c3b9 100644
> --- a/app/test-pmd/config.c
> +++ b/app/test-pmd/config.c
> @@ -947,6 +947,8 @@ static const struct {
>  	MK_FLOW_ITEM(MPLS, sizeof(struct rte_flow_item_mpls)),
>  	MK_FLOW_ITEM(GRE, sizeof(struct rte_flow_item_gre)),
>  	MK_FLOW_ITEM(FUZZY, sizeof(struct rte_flow_item_fuzzy)),

Remember to add GTP here assuming it makes sense.

> +	MK_FLOW_ITEM(GTPC, sizeof(struct rte_flow_item_gtp)),
> +	MK_FLOW_ITEM(GTPU, sizeof(struct rte_flow_item_gtp)),
>  };
>  
>  /** Compute storage space needed by item specification. */
> diff --git a/doc/guides/prog_guide/rte_flow.rst b/doc/guides/prog_guide/rte_flow.rst
> index 662a912..9e7179a 100644
> --- a/doc/guides/prog_guide/rte_flow.rst
> +++ b/doc/guides/prog_guide/rte_flow.rst
> @@ -955,6 +955,32 @@ Usage example, fuzzy match a TCPv4 packets:
>     | 4     | END      |
>     +-------+----------+
>  
> +Item: ``GTPC``
> +^^^^^^^^^^^^^^
> +
> +Matches a GTP header.
> +
> +- ``v_pt_rsv_flags``: version (3b), protocol type (1b), reserved (1b),
> +  extension header flag (1b), sequence number flag (1b), N-PDU number
> +  flag (1b).
> +- ``msg_type``: message type.
> +- ``msg_len``: message length.
> +- ``teid``: TEID.
> +- Default ``mask`` matches teid only.
> +
> +Item: ``GTPU``
> +^^^^^^^^^^^^^^
> +
> +Matches a GTP header.
> +
> +- ``v_pt_rsv_flags``: version (3b), protocol type (1b), reserved (1b),
> +  extension header flag (1b), sequence number flag (1b), N-PDU number
> +  flag (1b).
> +- ``msg_type``: message type.
> +- ``msg_len``: message length.
> +- ``teid``: TEID.
> +- Default ``mask`` matches teid only.
> +

You can use a single section to describe all three items at once since they
map to a common structure:

 Item: ``GTP``, ``GTPC``, ``GTPU``:
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Then elaborate a bit on the the differences between them.

>  Actions
>  ~~~~~~~
>  
> diff --git a/doc/guides/testpmd_app_ug/testpmd_funcs.rst b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
> index 2ed62f5..2ca36ad 100644
> --- a/doc/guides/testpmd_app_ug/testpmd_funcs.rst
> +++ b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
> @@ -2696,6 +2696,14 @@ This section lists supported pattern items and their attributes, if any.
>  
>    - ``thresh {unsigned}``: accuracy threshold.
>  
> +- ``gtpc``: match GTP header.
> +
> +  - ``teid {unsigned}``: Tunnel endpoint identifier.

Tunnel => tunnel

> +
> +- ``gtpu``: match GTP header.
> +
> +  - ``teid {unsigned}``: Tunnel endpoint identifier.

You could also merge all three items here, e.g.:

 - ``gtp``, ``gtpc``, ``gtpu``: ...

> +
>  Actions list
>  ^^^^^^^^^^^^
>  
> diff --git a/lib/librte_ether/rte_flow.h b/lib/librte_ether/rte_flow.h
> index bba6169..8b24cac 100644
> --- a/lib/librte_ether/rte_flow.h
> +++ b/lib/librte_ether/rte_flow.h
> @@ -309,6 +309,24 @@ enum rte_flow_item_type {
>  	 * See struct rte_flow_item_fuzzy.
>  	 */
>  	RTE_FLOW_ITEM_TYPE_FUZZY,
> +
> +	/**
> +	 * Matches a GTP header.

Write "GTP-U" to make clear this is not nonspecific "GTP" matching.

> +	 *
> +	 * Configure flow for GTP-C packets.
> +	 *
> +	 * See struct rte_flow_item_gtp.
> +	 */
> +	RTE_FLOW_ITEM_TYPE_GTPC,
> +
> +	/**
> +	 * Matches a GTP header.

"GTP-C" here.

> +	 *
> +	 * Configure flow for GTP-U packets.
> +	 *
> +	 * See struct rte_flow_item_gtp.
> +	 */
> +	RTE_FLOW_ITEM_TYPE_GTPU,
>  };
>  
>  /**
> @@ -735,6 +753,32 @@ static const struct rte_flow_item_fuzzy rte_flow_item_fuzzy_mask = {
>  #endif
>  
>  /**
> + * RTE_FLOW_ITEM_TYPE_GTP.

You need to mention the others, something like:

 RTE_FLOW_ITEM_TYPE_GTP, RTE_FLOW_ITEM_TYPE_GTPC and RTE_FLOW_ITEM_TYPE_GTPU.

> + *
> + * Matches a GTP header.

Similarly:

 Matches a nonspecific GTP, a GTP-C or a GTP-U header.

> + */
> +struct rte_flow_item_gtp {
> +	/**
> +	 * Version (2b), protocol type (1b), reserved (1b),
> +	 * Extension header flag (1b),
> +	 * Sequence number flag (1b),

Extension => extension
sequence => sequence

> +	 * N-PDU number flag (1b).
> +	 */
> +	uint8_t v_pt_rsv_flags;
> +	uint8_t msg_type; /**< Message type. */
> +	rte_be16_t msg_len; /**< Message length. */
> +	rte_be32_t teid; /**< Tunnel endpoint identifier. */
> +};
> +
> +/** Default mask for RTE_FLOW_ITEM_TYPE_GTP. */
> +#ifndef __cplusplus
> +static const struct rte_flow_item_gtp rte_flow_item_gtp_mask = {
> +	.msg_type = 0x00,

The above field is not necessary since you're not initializing the entire
structure, the rest is set to 0 by default.

> +	.teid = RTE_BE32(0xffffffff),
> +};
> +#endif
> +
> +/**
>   * Matching pattern item definition.
>   *
>   * A pattern is formed by stacking items starting from the lowest protocol
> -- 
> 2.5.5
>
  

Hi Adrien,

> -----Original Message-----
> From: Adrien Mazarguil [mailto:adrien.mazarguil@6wind.com]
> Sent: Thursday, September 7, 2017 8:20 PM
> To: Xing, Beilei <beilei.xing@intel.com>
> Cc: Wu, Jingjing <jingjing.wu@intel.com>; Chilikin, Andrey
> <andrey.chilikin@intel.com>; dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2 2/6] ethdev: add GTPC and GTPU items
> 
> Hi Beilei,
> 
> I assume this patch supersedes [1]?
> 
> [1] http://dpdk.org/ml/archives/dev/2017-August/073501.html
> 
> Thanks for merging testpmd and the API change as a single patch, I still have
> a few comments, see below.
> 
> (please add "flow API" somewhere in the title by the way)

Thanks for all your comments.
Yes, http://dpdk.org/ml/archives/dev/2017-August/073501.html is superseded and did merging testpmd and API change.
I will update title in next version.

> 
> On Thu, Sep 07, 2017 at 07:20:59PM +0800, Beilei Xing wrote:
> > This patch adds GTPC and GTPU items to generic rte flow, and also
> > exposes the following item fields through the flow command:
> >
> > - GTPC TEID
> > - GTPU TEID
> >
> > Signed-off-by: Beilei Xing <beilei.xing@intel.com>
> 
> Won't there be a need to match nonspecific GTP traffic as well (both GTP-C
> and GTP-U a once), since they use the same structure?
> 
> I'm not familiar with the protocol at all so I wonder if you should maybe leave
> the GTP item in addition to those two.
> 

Agree, I will leave the GTP item in next version.

GTP-C and GTP-U use the same structure, the difference between them is UDP port, 2123 is for GTP-C, and 2152 is for GTP-U.
Add GTP-C and GTP -U item since I want to design a user-friendly CLI.

For example, if user wants to add such flow: assign GTP-U packets with TEID 0x123456 to queue 14. 
Then use can use following CLI:
flow create 0 ingress pattern eth / ipv4 / udp / gtpu teid is 0x123456 / end actions queue index 14 / end
instead of below CLI to distinguish GTP-C and GTP-U with UDP port:
flow create 0 ingress pattern eth / ipv4 / udp dst spec 2125 dst mask 0 / gtp teid is 0x123456 / end actions queue index 14 / end

And all your other comments will be addressed in next version, thanks.

Beilei
  

Hi Beilei,

On Tue, Sep 12, 2017 at 06:40:50AM +0000, Xing, Beilei wrote:
> Hi Adrien,
> 
> > -----Original Message-----
> > From: Adrien Mazarguil [mailto:adrien.mazarguil@6wind.com]
> > Sent: Thursday, September 7, 2017 8:20 PM
> > To: Xing, Beilei <beilei.xing@intel.com>
> > Cc: Wu, Jingjing <jingjing.wu@intel.com>; Chilikin, Andrey
> > <andrey.chilikin@intel.com>; dev@dpdk.org
> > Subject: Re: [dpdk-dev] [PATCH v2 2/6] ethdev: add GTPC and GTPU items
> > 
> > Hi Beilei,
> > 
> > I assume this patch supersedes [1]?
> > 
> > [1] http://dpdk.org/ml/archives/dev/2017-August/073501.html
> > 
> > Thanks for merging testpmd and the API change as a single patch, I still have
> > a few comments, see below.
> > 
> > (please add "flow API" somewhere in the title by the way)
> 
> Thanks for all your comments.
> Yes, http://dpdk.org/ml/archives/dev/2017-August/073501.html is superseded and did merging testpmd and API change.
> I will update title in next version.

All right, thanks.

> > On Thu, Sep 07, 2017 at 07:20:59PM +0800, Beilei Xing wrote:
> > > This patch adds GTPC and GTPU items to generic rte flow, and also
> > > exposes the following item fields through the flow command:
> > >
> > > - GTPC TEID
> > > - GTPU TEID
> > >
> > > Signed-off-by: Beilei Xing <beilei.xing@intel.com>
> > 
> > Won't there be a need to match nonspecific GTP traffic as well (both GTP-C
> > and GTP-U a once), since they use the same structure?
> > 
> > I'm not familiar with the protocol at all so I wonder if you should maybe leave
> > the GTP item in addition to those two.
> > 
> 
> Agree, I will leave the GTP item in next version.
> 
> GTP-C and GTP-U use the same structure, the difference between them is UDP port, 2123 is for GTP-C, and 2152 is for GTP-U.
> Add GTP-C and GTP -U item since I want to design a user-friendly CLI.
> 
> For example, if user wants to add such flow: assign GTP-U packets with TEID 0x123456 to queue 14. 
> Then use can use following CLI:
> flow create 0 ingress pattern eth / ipv4 / udp / gtpu teid is 0x123456 / end actions queue index 14 / end
> instead of below CLI to distinguish GTP-C and GTP-U with UDP port:
> flow create 0 ingress pattern eth / ipv4 / udp dst spec 2125 dst mask 0 / gtp teid is 0x123456 / end actions queue index 14 / end

I agree with you from a usability standpoint it's much nicer. I have one
question though, could one send GTP-C / GTP-U traffic using nondefault UDP
ports and expect hardware to match it without explicitly specifying a port
in the UDP item, that is, is there some property in GTP-C / GTP-U traffic
outside the UDP port that would theoretically allow a host (even a stateful
one) to tell them apart?

If it really depends on the UDP port only, not specifying one will use
hardware defaults regardless of the item (GTP / GTP-U / GTP-C). However if
like VXLAN, this default value can be modified outside of rte_flow, most
users will have to specify it regardless in order to get consistent results
across various vendors/adapters.

In any case I don't mind three items. GTP encompasses both GTP-U and GTP-C
(possibly two different UDP ports at once), while GTP-U and GTP-C match
exactly one. You only have to describe this properly in the documentation.

Thanks.
  

Hi Adrien,

> -----Original Message-----
> From: Adrien Mazarguil [mailto:adrien.mazarguil@6wind.com]
> Sent: Tuesday, September 12, 2017 6:47 PM
> To: Xing, Beilei <beilei.xing@intel.com>
> Cc: Wu, Jingjing <jingjing.wu@intel.com>; Chilikin, Andrey
> <andrey.chilikin@intel.com>; dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2 2/6] ethdev: add GTPC and GTPU items
> 
> Hi Beilei,
> 
> On Tue, Sep 12, 2017 at 06:40:50AM +0000, Xing, Beilei wrote:
> > Hi Adrien,
> >
> > > -----Original Message-----
> > > From: Adrien Mazarguil [mailto:adrien.mazarguil@6wind.com]
> > > Sent: Thursday, September 7, 2017 8:20 PM
> > > To: Xing, Beilei <beilei.xing@intel.com>
> > > Cc: Wu, Jingjing <jingjing.wu@intel.com>; Chilikin, Andrey
> > > <andrey.chilikin@intel.com>; dev@dpdk.org
> > > Subject: Re: [dpdk-dev] [PATCH v2 2/6] ethdev: add GTPC and GTPU
> > > items
> > >
> > > Hi Beilei,
> > >
> > > I assume this patch supersedes [1]?
> > >
> > > [1] http://dpdk.org/ml/archives/dev/2017-August/073501.html
> > >
> > > Thanks for merging testpmd and the API change as a single patch, I
> > > still have a few comments, see below.
> > >
> > > (please add "flow API" somewhere in the title by the way)
> >
> > Thanks for all your comments.
> > Yes, http://dpdk.org/ml/archives/dev/2017-August/073501.html is
> superseded and did merging testpmd and API change.
> > I will update title in next version.
> 
> All right, thanks.
> 
> > > On Thu, Sep 07, 2017 at 07:20:59PM +0800, Beilei Xing wrote:
> > > > This patch adds GTPC and GTPU items to generic rte flow, and also
> > > > exposes the following item fields through the flow command:
> > > >
> > > > - GTPC TEID
> > > > - GTPU TEID
> > > >
> > > > Signed-off-by: Beilei Xing <beilei.xing@intel.com>
> > >
> > > Won't there be a need to match nonspecific GTP traffic as well (both
> > > GTP-C and GTP-U a once), since they use the same structure?
> > >
> > > I'm not familiar with the protocol at all so I wonder if you should
> > > maybe leave the GTP item in addition to those two.
> > >
> >
> > Agree, I will leave the GTP item in next version.
> >
> > GTP-C and GTP-U use the same structure, the difference between them is
> UDP port, 2123 is for GTP-C, and 2152 is for GTP-U.
> > Add GTP-C and GTP -U item since I want to design a user-friendly CLI.
> >
> > For example, if user wants to add such flow: assign GTP-U packets with
> TEID 0x123456 to queue 14.
> > Then use can use following CLI:
> > flow create 0 ingress pattern eth / ipv4 / udp / gtpu teid is 0x123456
> > / end actions queue index 14 / end instead of below CLI to distinguish
> GTP-C and GTP-U with UDP port:
> > flow create 0 ingress pattern eth / ipv4 / udp dst spec 2125 dst mask
> > 0 / gtp teid is 0x123456 / end actions queue index 14 / end
> 
> I agree with you from a usability standpoint it's much nicer. I have one
> question though, could one send GTP-C / GTP-U traffic using nondefault UDP
> ports and expect hardware to match it without explicitly specifying a port in
> the UDP item, that is, is there some property in GTP-C / GTP-U traffic outside
> the UDP port that would theoretically allow a host (even a stateful
> one) to tell them apart?
> 
> If it really depends on the UDP port only, not specifying one will use
> hardware defaults regardless of the item (GTP / GTP-U / GTP-C). However if
> like VXLAN, this default value can be modified outside of rte_flow, most
> users will have to specify it regardless in order to get consistent results
> across various vendors/adapters.

As far as I know, there's no other property in GTP-C / GTP-U traffic outside the UDP port to distinguish GTP-C and GTP-U. And I don't think the value can be modified by user just like VXLAN.

> 
> In any case I don't mind three items. GTP encompasses both GTP-U and
> GTP-C (possibly two different UDP ports at once), while GTP-U and GTP-C
> match exactly one. You only have to describe this properly in the
> documentation.

OK, I will describe it in documentation.

Beilei