Skip to contentSkip to conversation

ethdev: introduce GENEVE header TLV option item

PATCH Superseded Shiri Kuzin 5 participants 57 messages Jan 7, 2021 – Jan 18, 2021 1 message not in the archive Network drivers

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

#1 lore raw

v3 cover ethdev: introduce GENEVE header TLV option item

The Geneve tunneling protocol is designed to allow the user to specify some data context on the packet.
The GENEVE TLV (Type-Length-Variable) Option is the mean intended to present the user data.

In order to support GENEVE TLV Option the new rte_flow item "rte_flow_item_geneve_opt" is introduced.
The new item contains the values and masks for the following fields:
-option class
-option type
-length
-data

The usage example:
"flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 / geneve-opt class is 5 length is 1 type is 0 data is 0x66998800 / end actions count / drop / end"

New item will be added to testpmd to support raw encap/decap action.

Shiri Kuzin (7):
lib/librte_ethdev: introduce GENEVE header TLV option item common/mlx5: check GENEVE TLV support in HCA attributes common/mlx5: create GENEVE TLV option object with DevX net/mlx5: create GENEVE TLV option management net/mlx5: add GENEVE TLV option flow validation net/mlx5: add GENEVE TLV option flow translation
doc: update GENEVE TLV option support

Viacheslav Ovsiienko (1):
app/testpmd: add GENEVE option item support

 app/test-pmd/cmdline_flow.c                 | 102 ++++++++++-
 doc/guides/nics/mlx5.rst                    |  23 ++-
 doc/guides/rel_notes/release_21_02.rst      |   8 +
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |   8 +
 drivers/common/mlx5/mlx5_devx_cmds.c        |  63 ++++++-
 drivers/common/mlx5/mlx5_devx_cmds.h        |   9 +
 drivers/common/mlx5/mlx5_prm.h              |  28 ++-
 drivers/common/mlx5/version.map             |   1 +
 drivers/net/mlx5/mlx5.c                     |   2 +
 drivers/net/mlx5/mlx5.h                     |  13 ++
 drivers/net/mlx5/mlx5_flow.c                | 120 +++++++++++++
 drivers/net/mlx5/mlx5_flow.h                |  11 ++
 drivers/net/mlx5/mlx5_flow_dv.c             | 188 +++++++++++++++++++-
 lib/librte_ethdev/rte_flow.c                |   1 +
 lib/librte_ethdev/rte_flow.h                |  27 +++
 15 files changed, 592 insertions(+), 12 deletions(-)
Signature · 1 line
2.21.0
#2 lore raw

PATCH v3 1/8 lib/librte_ethdev: introduce GENEVE header TLV option item

The Geneve tunneling protocol is designed to allow the user to specify some data context on the packet. The GENEVE TLV (Type-Length-Variable) Option is the mean intended to present the user data.

In order to support GENEVE TLV Option the new rte_flow item "rte_flow_item_geneve_opt" is added.
The new item contains the values and masks for the following fields:
-option class
-option type
-length
-data

New item will be added to testpmd to support match and raw encap/decap actions.

---
 lib/librte_ethdev/rte_flow.c |  1 +
 lib/librte_ethdev/rte_flow.h | 27 +++++++++++++++++++++++++++
 2 files changed, 28 insertions(+)
lib/librte_ethdev/rte_flow.c +1 -0
index a06f64c271..2af7d965e1 100644
@@ -97,6 +97,7 @@ static const struct rte_flow_desc_data rte_flow_desc_item[] = {
 	MK_FLOW_ITEM(L2TPV3OIP, sizeof(struct rte_flow_item_l2tpv3oip)),
 	MK_FLOW_ITEM(PFCP, sizeof(struct rte_flow_item_pfcp)),
 	MK_FLOW_ITEM(ECPRI, sizeof(struct rte_flow_item_ecpri)),
+	MK_FLOW_ITEM(GENEVE_OPT, sizeof(struct rte_flow_item_geneve_opt)),
 };
 
 /** Generate flow_action[] entry. */
lib/librte_ethdev/rte_flow.h +27 -0
index 0977a78270..11a6494b8e 100644
@@ -543,6 +543,14 @@ enum rte_flow_item_type {
 	 * See struct rte_flow_item_ipv6_frag_ext.
 	 */
 	RTE_FLOW_ITEM_TYPE_IPV6_FRAG_EXT,
+
+	/**
+	 * Matches Geneve Variable Length Option
+	 *
+	 * See struct rte_flow_item_geneve_opt
+	 */
+	RTE_FLOW_ITEM_TYPE_GENEVE_OPT,
+
 };
 
 /**
@@ -1627,6 +1635,25 @@ static const struct rte_flow_item_ecpri rte_flow_item_ecpri_mask = {
 };
 #endif
 
+/**
+ * RTE_FLOW_ITEM_TYPE_GENEVE_OPT
+ *
+ * Matches a GENEVE Variable Length Option
+ */
+struct rte_flow_item_geneve_opt {
+	rte_be16_t option_class;
+	uint8_t option_type;
+	uint8_t option_len;
+	uint32_t *data;
+};
+
+/** Default mask for RTE_FLOW_ITEM_TYPE_GENEVE_OPT. */
+#ifndef __cplusplus
+static const struct rte_flow_item_geneve_opt
+rte_flow_item_geneve_opt_mask = {
+	.option_type = 0xff,
+};
+#endif
 /**
  * Matching pattern item definition.
  *
Signature · 1 line
2.21.0
Acked-by on v3 1/8
#3 lore raw

Re: [dpdk-dev] [PATCH v3 1/8] lib/librte_ethdev: introduce GENEVE header TLV option item

Hi,

Quoted 88 lines

-----Original Message-----
From: Shiri Kuzin <[email protected]>
Sent: Thursday, January 7, 2021 10:39 AM
Subject: [PATCH v3 1/8] lib/librte_ethdev: introduce GENEVE header TLV option item

The Geneve tunneling protocol is designed to allow the user to specify some data context on the packet. The GENEVE TLV (Type-Length-Variable) Option is the mean intended to present the user data.

In order to support GENEVE TLV Option the new rte_flow item "rte_flow_item_geneve_opt" is added.
The new item contains the values and masks for the following fields:

(quoted diff) +0 -4
-option class
-option type
-length
-data

New item will be added to testpmd to support match and raw encap/decap actions.

---
 lib/librte_ethdev/rte_flow.c |  1 +
 lib/librte_ethdev/rte_flow.h | 27 +++++++++++++++++++++++++++
 2 files changed, 28 insertions(+)
lib/librte_ethdev/rte_flow.c +1 -0
index a06f64c271..2af7d965e1 100644
@@ -97,6 +97,7 @@ static const struct rte_flow_desc_data
 rte_flow_desc_item[] = {
 	MK_FLOW_ITEM(L2TPV3OIP, sizeof(struct rte_flow_item_l2tpv3oip)),
 	MK_FLOW_ITEM(PFCP, sizeof(struct rte_flow_item_pfcp)),
 	MK_FLOW_ITEM(ECPRI, sizeof(struct rte_flow_item_ecpri)),
+	MK_FLOW_ITEM(GENEVE_OPT, sizeof(struct
 rte_flow_item_geneve_opt)),
 };

/** Generate flow_action[] entry. */

lib/librte_ethdev/rte_flow.h +27 -0
index 0977a78270..11a6494b8e 100644
@@ -543,6 +543,14 @@ enum rte_flow_item_type {
 	 * See struct rte_flow_item_ipv6_frag_ext.
 	 */
 	RTE_FLOW_ITEM_TYPE_IPV6_FRAG_EXT,
+
+	/**
+	 * Matches Geneve Variable Length Option
+	 *
+	 * See struct rte_flow_item_geneve_opt
+	 */
+	RTE_FLOW_ITEM_TYPE_GENEVE_OPT,
+
 };
 
 /**
@@ -1627,6 +1635,25 @@ static const struct rte_flow_item_ecpri
 rte_flow_item_ecpri_mask = {
 };
 #endif
 
+/**
+ * RTE_FLOW_ITEM_TYPE_GENEVE_OPT
+ *
+ * Matches a GENEVE Variable Length Option
+ */
+struct rte_flow_item_geneve_opt {
+	rte_be16_t option_class;
+	uint8_t option_type;
+	uint8_t option_len;
+	uint32_t *data;
+};
+
+/** Default mask for RTE_FLOW_ITEM_TYPE_GENEVE_OPT. */
+#ifndef __cplusplus
+static const struct rte_flow_item_geneve_opt
+rte_flow_item_geneve_opt_mask = {
+	.option_type = 0xff,
+};
+#endif
 /**
  * Matching pattern item definition.
Quoted 3 lines

*
--
2.21.0

Thanks,
Ori

#4 lore raw

PATCH v3 2/8 app/testpmd: add GENEVE option item support

In reply to Shiri Kuzin

From: Viacheslav Ovsiienko <[email protected]>

The patch adds the GENEVE option rte flow item support to command line interpreter. The flow command with GENEVE option items looks like:

flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 /
geneve-opt class is 99 length is 1 type is 0 data is 0x669988 / end actions drop / end

The option length should be specified in 32-bit words, this value specifies the length of the data pattern/mask arrays (should be multiplied by sizeof(uint32_t) to be expressed in bytes. If match on the length itself is not needed the mask should be set to zero, in this case length is used to specify the pattern/mask array lengths only.

---
 app/test-pmd/cmdline_flow.c                 | 102 ++++++++++++++++++--
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |   8 ++
 2 files changed, 104 insertions(+), 6 deletions(-)
app/test-pmd/cmdline_flow.c +96 -6
index 585cab98b4..8bb2cb1b2d 100644
@@ -283,6 +283,11 @@ enum index {
 	ITEM_ECPRI_MSG_IQ_DATA_PCID,
 	ITEM_ECPRI_MSG_RTC_CTRL_RTCID,
 	ITEM_ECPRI_MSG_DLY_MSR_MSRID,
+	ITEM_GENEVE_OPT,
+	ITEM_GENEVE_OPT_CLASS,
+	ITEM_GENEVE_OPT_TYPE,
+	ITEM_GENEVE_OPT_LENGTH,
+	ITEM_GENEVE_OPT_DATA,
 
 	/* Validate/create actions. */
 	ACTIONS,
@@ -413,6 +418,9 @@ enum index {
 /** Maximum size for pattern in struct rte_flow_item_raw. */
 #define ITEM_RAW_PATTERN_SIZE 40
 
+/** Maximum size for GENEVE option data pattern in bytes. */
+#define ITEM_GENEVE_OPT_DATA_SIZE 124
+
 /** Storage size for struct rte_flow_item_raw including pattern. */
 #define ITEM_RAW_SIZE \
 	(sizeof(struct rte_flow_item_raw) + ITEM_RAW_PATTERN_SIZE)
@@ -428,7 +436,7 @@ struct action_rss_data {
 };
 
 /** Maximum data size in struct rte_flow_action_raw_encap. */
-#define ACTION_RAW_ENCAP_MAX_DATA 128
+#define ACTION_RAW_ENCAP_MAX_DATA 512
 #define RAW_ENCAP_CONFS_MAX_NUM 8
 
 /** Storage for struct rte_flow_action_raw_encap. */
@@ -658,6 +666,16 @@ struct token {
 		.mask = (const void *)&(const s){ .f = (1 << (b)) - 1 }, \
 	})
 
+/** Static initializer for ARGS() to target a field with limits. */
+#define ARGS_ENTRY_BOUNDED(s, f, i, a) \
+	(&(const struct arg){ \
+		.bounded = 1, \
+		.min = (i), \
+		.max = (a), \
+		.offset = offsetof(s, f), \
+		.size = sizeof(((s *)0)->f), \
+	})
+
 /** Static initializer for ARGS() to target an arbitrary bit-mask. */
 #define ARGS_ENTRY_MASK(s, f, m) \
 	(&(const struct arg){ \
@@ -903,6 +921,7 @@ static const enum index next_item[] = {
 	ITEM_AH,
 	ITEM_PFCP,
 	ITEM_ECPRI,
+	ITEM_GENEVE_OPT,
 	END_SET,
 	ZERO,
 };
@@ -1244,6 +1263,15 @@ static const enum index item_ecpri_common_type[] = {
 	ZERO,
 };
 
+static const enum index item_geneve_opt[] = {
+	ITEM_GENEVE_OPT_CLASS,
+	ITEM_GENEVE_OPT_TYPE,
+	ITEM_GENEVE_OPT_LENGTH,
+	ITEM_GENEVE_OPT_DATA,
+	ITEM_NEXT,
+	ZERO,
+};
+
 static const enum index next_action[] = {
 	ACTION_END,
 	ACTION_VOID,
@@ -3230,6 +3258,47 @@ static const struct token token_list[] = {
 		.args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_ecpri,
 				hdr.type5.msr_id)),
 	},
+	[ITEM_GENEVE_OPT] = {
+		.name = "geneve-opt",
+		.help = "GENEVE header option",
+		.priv = PRIV_ITEM(GENEVE_OPT,
+				  sizeof(struct rte_flow_item_geneve_opt) +
+				  ITEM_GENEVE_OPT_DATA_SIZE),
+		.next = NEXT(item_geneve_opt),
+		.call = parse_vc,
+	},
+	[ITEM_GENEVE_OPT_CLASS]	= {
+		.name = "class",
+		.help = "GENEVE option class",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(UNSIGNED), item_param),
+		.args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_geneve_opt,
+					     option_class)),
+	},
+	[ITEM_GENEVE_OPT_TYPE] = {
+		.name = "type",
+		.help = "GENEVE option type",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(UNSIGNED), item_param),
+		.args = ARGS(ARGS_ENTRY(struct rte_flow_item_geneve_opt,
+					option_type)),
+	},
+	[ITEM_GENEVE_OPT_LENGTH] = {
+		.name = "length",
+		.help = "GENEVE option data length (in 32b words)",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(UNSIGNED), item_param),
+		.args = ARGS(ARGS_ENTRY_BOUNDED(
+				struct rte_flow_item_geneve_opt, option_len,
+				0, 31)),
+	},
+	[ITEM_GENEVE_OPT_DATA] = {
+		.name = "data",
+		.help = "GENEVE option data pattern",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(HEX), item_param),
+		.args = ARGS(ARGS_ENTRY(struct rte_flow_item_geneve_opt, data),
+			     ARGS_ENTRY_ARB(0, 0),
+			     ARGS_ENTRY_ARB
+				(sizeof(struct rte_flow_item_geneve_opt),
+				ITEM_GENEVE_OPT_DATA_SIZE)),
+	},
 	/* Validate/create actions. */
 	[ACTIONS] = {
 		.name = "actions",
@@ -6482,11 +6551,14 @@ parse_hex(struct context *ctx, const struct token *token,
 	ret = snprintf(tmp, sizeof(tmp), "%u", hexlen);
 	if (ret < 0)
 		goto error;
-	push_args(ctx, arg_len);
-	ret = parse_int(ctx, token, tmp, ret, NULL, 0);
-	if (ret < 0) {
-		pop_args(ctx);
-		goto error;
+	/* Save length if requested. */
+	if (arg_len->size) {
+		push_args(ctx, arg_len);
+		ret = parse_int(ctx, token, tmp, ret, NULL, 0);
+		if (ret < 0) {
+			pop_args(ctx);
+			goto error;
+		}
 	}
 	buf = (uint8_t *)ctx->object + arg_data->offset;
 	/* Output buffer is not necessarily NUL-terminated. */
@@ -7486,6 +7558,9 @@ flow_item_default_mask(const struct rte_flow_item *item)
 	case RTE_FLOW_ITEM_TYPE_GENEVE:
 		mask = &rte_flow_item_geneve_mask;
 		break;
+	case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+		mask = &rte_flow_item_geneve_opt_mask;
+		break;
 	case RTE_FLOW_ITEM_TYPE_PPPOE_PROTO_ID:
 		mask = &rte_flow_item_pppoe_proto_id_mask;
 		break;
@@ -7598,6 +7673,8 @@ cmd_set_raw_parsed(const struct buffer *in)
 	/* process hdr from upper layer to low layer (L3/L4 -> L2). */
 	data_tail = data + ACTION_RAW_ENCAP_MAX_DATA;
 	for (i = n - 1 ; i >= 0; --i) {
+		const struct rte_flow_item_geneve_opt *opt;
+
 		item = in->args.vc.pattern + i;
 		if (item->spec == NULL)
 			item->spec = flow_item_default_mask(item);
@@ -7650,6 +7727,19 @@ cmd_set_raw_parsed(const struct buffer *in)
 		case RTE_FLOW_ITEM_TYPE_GENEVE:
 			size = sizeof(struct rte_geneve_hdr);
 			break;
+		case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+			opt = (const struct rte_flow_item_geneve_opt *)
+								item->spec;
+			size = sizeof(struct rte_flow_item_geneve_opt) -
+			       sizeof(opt->data);
+			if (opt->option_len && opt->data) {
+				*total_size += opt->option_len *
+					       sizeof(uint32_t);
+				rte_memcpy(data_tail - (*total_size),
+					   opt->data,
+					   opt->option_len * sizeof(uint32_t));
+			}
+			break;
 		case RTE_FLOW_ITEM_TYPE_L2TPV3OIP:
 			size = sizeof(rte_be32_t);
 			proto = 0x73;
doc/guides/testpmd_app_ug/testpmd_funcs.rst +8 -0
index 9be450066e..37278d31d6 100644
@@ -3680,6 +3680,14 @@ This section lists supported pattern items and their attributes, if any.
   - ``vni {unsigned}``: virtual network identifier.
   - ``protocol {unsigned}``: protocol type.
 
+- ``geneve-opt``: match GENEVE header option.
+
+  - ``class {unsigned}``: GENEVE option class.
+  - ``type {unsigned}``: GENEVE option type.
+  - ``length {unsigned}``: GENEVE option length in 32-bit words.
+  - ``data {hex string}``: GENEVE option data, the legnt is defined by
+    ``length`` field
+
 - ``vxlan-gpe``: match VXLAN-GPE header.
 
   - ``vni {unsigned}``: VXLAN-GPE identifier.
Signature · 1 line
2.21.0
#5 lore raw

Re: [dpdk-dev] [PATCH v3 2/8] app/testpmd: add GENEVE option item support

Hi,

Quoted 32 lines

-----Original Message-----
From: dev <[email protected]> On Behalf Of Shiri Kuzin
Sent: Thursday, January 7, 2021 4:39 PM
To: [email protected]
Cc: Slava Ovsiienko <[email protected]>; NBU-Contact-Adrien Mazarguil <[email protected]>; Ori Kam <[email protected]>; [email protected]; NBU-Contact-Thomas Monjalon <[email protected]>; Raslan Darawsheh <[email protected]>
Subject: [dpdk-dev] [PATCH v3 2/8] app/testpmd: add GENEVE option item support

From: Viacheslav Ovsiienko <[email protected]>

The patch adds the GENEVE option rte flow item support to command line interpreter. The flow command with GENEVE option items looks like:

flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 /
geneve-opt class is 99 length is 1 type is 0 data is 0x669988 / end actions drop / end

The option length should be specified in 32-bit words, this value specifies the length of the data pattern/mask arrays (should be multiplied by sizeof(uint32_t) to be expressed in bytes. If match on the length itself is not needed the mask should be set to zero, in this case length is used to specify the pattern/mask array lengths only.

---
 app/test-pmd/cmdline_flow.c                 | 102 ++++++++++++++++++--
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |   8 ++
 2 files changed, 104 insertions(+), 6 deletions(-)

snip

Quoted 21 lines
app/test-pmd/cmdline_flow.c +6 -0
 	/* Output buffer is not necessarily NUL-terminated. */ @@ -7486,6
+7558,9 @@ flow_item_default_mask(const struct rte_flow_item *item)
 	case RTE_FLOW_ITEM_TYPE_GENEVE:
 		mask = &rte_flow_item_geneve_mask;
 		break;
+	case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+		mask = &rte_flow_item_geneve_opt_mask;
+		break;
 	case RTE_FLOW_ITEM_TYPE_PPPOE_PROTO_ID:
 		mask = &rte_flow_item_pppoe_proto_id_mask;
 		break;
@@ -7598,6 +7673,8 @@ cmd_set_raw_parsed(const struct buffer *in)
 	/* process hdr from upper layer to low layer (L3/L4 -> L2). */
 	data_tail = data + ACTION_RAW_ENCAP_MAX_DATA;
 	for (i = n - 1 ; i >= 0; --i) {
+		const struct rte_flow_item_geneve_opt *opt;
+
 		item = in->args.vc.pattern + i;
 		if (item->spec == NULL)
 			item->spec = flow_item_default_mask(item); @@ -

7650,6 +7727,19 @@ cmd_set_raw_parsed(const struct buffer *in)

app/test-pmd/cmdline_flow.c cmd_set_raw_parsed(const struct buffer *in) · line 7734

Quoted 8 lines
app/test-pmd/cmdline_flow.c +5 -0
 		case RTE_FLOW_ITEM_TYPE_GENEVE:
 			size = sizeof(struct rte_geneve_hdr);
 			break;
+		case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+			opt = (const struct rte_flow_item_geneve_opt *)
+								item->spec;
+			size = sizeof(struct rte_flow_item_geneve_opt) -
+			       sizeof(opt->data);

If I understand correctly, the potential here is to calculate the geneve option size, right? Have we confirmed the struct is well aligned for all arch, and will not be aligned to other size? Or is it better to use "offsetof" here?

Quoted 34 lines
app/test-pmd/cmdline_flow.c +8 -0
+			if (opt->option_len && opt->data) {
+				*total_size += opt->option_len *
+					       sizeof(uint32_t);
+				rte_memcpy(data_tail - (*total_size),
+					   opt->data,
+					   opt->option_len * sizeof(uint32_t));
+			}
+			break;
 		case RTE_FLOW_ITEM_TYPE_L2TPV3OIP:
 			size = sizeof(rte_be32_t);
 			proto = 0x73;
doc/guides/testpmd_app_ug/testpmd_funcs.rst +0 -0
index 9be450066e..37278d31d6 100644
@@ -3680,6 +3680,14 @@ This section lists supported pattern items and their

attributes, if any.
- ``vni {unsigned}``: virtual network identifier.
- ``protocol {unsigned}``: protocol type.

+- ``geneve-opt``: match GENEVE header option.

doc/guides/testpmd_app_ug/testpmd_funcs.rst +7 -1
+
+  - ``class {unsigned}``: GENEVE option class.
+  - ``type {unsigned}``: GENEVE option type.
+  - ``length {unsigned}``: GENEVE option length in 32-bit words.
+  - ``data {hex string}``: GENEVE option data, the legnt is defined by
+    ``length`` field
+
 - ``vxlan-gpe``: match VXLAN-GPE header.
 
   - ``vni {unsigned}``: VXLAN-GPE identifier.
--

2.21.0

#6 lore raw

Re: [dpdk-dev] [PATCH v3 2/8] app/testpmd: add GENEVE option item support

Hi,

Quoted 75 lines

-----Original Message-----
From: Suanming Mou <[email protected]>
Sent: Friday, January 8, 2021 14:11
To: Shiri Kuzin <[email protected]>; [email protected]
Cc: Slava Ovsiienko <[email protected]>; NBU-Contact-Adrien Mazarguil <[email protected]>; Ori Kam <[email protected]>; [email protected]; NBU-Contact-Thomas Monjalon <[email protected]>; Raslan Darawsheh <[email protected]>
Subject: RE: [dpdk-dev] [PATCH v3 2/8] app/testpmd: add GENEVE option item support

Hi,

Quoted 32 lines

-----Original Message-----
From: dev <[email protected]> On Behalf Of Shiri Kuzin
Sent: Thursday, January 7, 2021 4:39 PM
To: [email protected]
Cc: Slava Ovsiienko <[email protected]>; NBU-Contact-Adrien Mazarguil <[email protected]>; Ori Kam <[email protected]>; [email protected]; NBU-Contact-Thomas Monjalon <[email protected]>; Raslan Darawsheh <[email protected]>
Subject: [dpdk-dev] [PATCH v3 2/8] app/testpmd: add GENEVE option item support

From: Viacheslav Ovsiienko <[email protected]>

The patch adds the GENEVE option rte flow item support to command line interpreter. The flow command with GENEVE option items looks like:

flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 /
geneve-opt class is 99 length is 1 type is 0 data is 0x669988 / end actions drop / end

The option length should be specified in 32-bit words, this value specifies the length of the data pattern/mask arrays (should be multiplied by sizeof(uint32_t) to be expressed in bytes. If match on the length itself is not needed the mask should be set to zero, in this case length is used to specify the pattern/mask array lengths only.

---
 app/test-pmd/cmdline_flow.c                 | 102 ++++++++++++++++++--
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |   8 ++
 2 files changed, 104 insertions(+), 6 deletions(-)

snip

Quoted 27 lines
app/test-pmd/cmdline_flow.c +6 -0
 	/* Output buffer is not necessarily NUL-terminated. */ @@ -7486,6
+7558,9 @@ flow_item_default_mask(const struct rte_flow_item *item)
 	case RTE_FLOW_ITEM_TYPE_GENEVE:
 		mask = &rte_flow_item_geneve_mask;
 		break;
+	case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+		mask = &rte_flow_item_geneve_opt_mask;
+		break;
 	case RTE_FLOW_ITEM_TYPE_PPPOE_PROTO_ID:
 		mask = &rte_flow_item_pppoe_proto_id_mask;
 		break;
@@ -7598,6 +7673,8 @@ cmd_set_raw_parsed(const struct buffer *in)
 	/* process hdr from upper layer to low layer (L3/L4 -> L2). */
 	data_tail = data + ACTION_RAW_ENCAP_MAX_DATA;
 	for (i = n - 1 ; i >= 0; --i) {
+		const struct rte_flow_item_geneve_opt *opt;
+
 		item = in->args.vc.pattern + i;
 		if (item->spec == NULL)
 			item->spec = flow_item_default_mask(item); @@ -

7650,6 +7727,19 @@ cmd_set_raw_parsed(const struct buffer *in)

app/test-pmd/cmdline_flow.c +3 -0
 		case RTE_FLOW_ITEM_TYPE_GENEVE:
 			size = sizeof(struct rte_geneve_hdr);
 			break;
+		case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+			opt = (const struct rte_flow_item_geneve_opt *)
+								item->spec;
Quoted 8 lines

app/test-pmd/cmdline_flow.c cmd_set_raw_parsed(const struct buffer *in) · line 7734

Quoted 2 lines
app/test-pmd/cmdline_flow.c +2 -0
+			size = sizeof(struct rte_flow_item_geneve_opt) -
+			       sizeof(opt->data);

If I understand correctly, the potential here is to calculate the geneve option size, right?
Have we confirmed the struct is well aligned for all arch, and will not be aligned to other size?
Or is it better to use "offsetof" here?

Yes, you are right, "offsetof" would be more relevant, thank you the catching,

With best regards,
Slava

Quoted 35 lines
Quoted 35 lines
app/test-pmd/cmdline_flow.c +8 -0
+			if (opt->option_len && opt->data) {
+				*total_size += opt->option_len *
+					       sizeof(uint32_t);
+				rte_memcpy(data_tail - (*total_size),
+					   opt->data,
+					   opt->option_len * sizeof(uint32_t));
+			}
+			break;
 		case RTE_FLOW_ITEM_TYPE_L2TPV3OIP:
 			size = sizeof(rte_be32_t);
 			proto = 0x73;
doc/guides/testpmd_app_ug/testpmd_funcs.rst +0 -0
index 9be450066e..37278d31d6 100644
@@ -3680,6 +3680,14 @@ This section lists supported pattern items and

their attributes, if any.
- ``vni {unsigned}``: virtual network identifier.
- ``protocol {unsigned}``: protocol type.

+- ``geneve-opt``: match GENEVE header option.

doc/guides/testpmd_app_ug/testpmd_funcs.rst +7 -1
+
+  - ``class {unsigned}``: GENEVE option class.
+  - ``type {unsigned}``: GENEVE option type.
+  - ``length {unsigned}``: GENEVE option length in 32-bit words.
+  - ``data {hex string}``: GENEVE option data, the legnt is defined by
+    ``length`` field
+
 - ``vxlan-gpe``: match VXLAN-GPE header.
 
   - ``vni {unsigned}``: VXLAN-GPE identifier.
--

2.21.0

#7 lore raw

PATCH v3 3/8 common/mlx5: check GENEVE TLV support in HCA attributes

In reply to Shiri Kuzin

This is preparation step to support match on GENEVE TLV option.

In this Patch we add the HCA attributes that will allow supporting GENEVE TLV option matching.

---
 drivers/common/mlx5/mlx5_devx_cmds.c |  7 +++++++
 drivers/common/mlx5/mlx5_devx_cmds.h |  4 ++++
 drivers/common/mlx5/mlx5_prm.h       | 28 +++++++++++++++++++++++++---
 3 files changed, 36 insertions(+), 3 deletions(-)
drivers/common/mlx5/mlx5_devx_cmds.c +7 -0
index 12f51a940c..fc406af062 100644
@@ -693,6 +693,10 @@ mlx5_devx_cmd_query_hca_attr(void *ctx,
 	attr->eth_virt = MLX5_GET(cmd_hca_cap, hcattr, eth_virt);
 	attr->flex_parser_protocols = MLX5_GET(cmd_hca_cap, hcattr,
 					       flex_parser_protocols);
+	attr->max_geneve_tlv_options = MLX5_GET(cmd_hca_cap, hcattr,
+			max_geneve_tlv_options);
+	attr->max_geneve_tlv_option_data_len = MLX5_GET(cmd_hca_cap, hcattr,
+			max_geneve_tlv_option_data_len);
 	attr->qos.sup = MLX5_GET(cmd_hca_cap, hcattr, qos);
 	attr->vdpa.valid = !!(MLX5_GET64(cmd_hca_cap, hcattr,
 					 general_obj_types) &
@@ -720,6 +724,9 @@ mlx5_devx_cmd_query_hca_attr(void *ctx,
 	attr->flow_hit_aso = !!(MLX5_GET64(cmd_hca_cap, hcattr,
 					   general_obj_types) &
 				MLX5_GENERAL_OBJ_TYPES_CAP_FLOW_HIT_ASO);
+	attr->geneve_tlv_opt = !!(MLX5_GET64(cmd_hca_cap, hcattr,
+					   general_obj_types) &
+				MLX5_GENERAL_OBJ_TYPES_CAP_GENEVE_TLV_OPT);
 	attr->log_max_cq = MLX5_GET(cmd_hca_cap, hcattr, log_max_cq);
 	attr->log_max_qp = MLX5_GET(cmd_hca_cap, hcattr, log_max_qp);
 	attr->log_max_cq_sz = MLX5_GET(cmd_hca_cap, hcattr, log_max_cq_sz);
drivers/common/mlx5/mlx5_devx_cmds.h +4 -0
index b335b7c82c..6b2f77837b 100644
@@ -97,6 +97,8 @@ struct mlx5_hca_attr {
 	uint32_t lro_timer_supported_periods[MLX5_LRO_NUM_SUPP_PERIODS];
 	uint16_t lro_min_mss_size;
 	uint32_t flex_parser_protocols;
+	uint32_t max_geneve_tlv_options;
+	uint32_t max_geneve_tlv_option_data_len;
 	uint32_t hairpin:1;
 	uint32_t log_max_hairpin_queues:5;
 	uint32_t log_max_hairpin_wq_data_sz:5;
@@ -116,6 +118,7 @@ struct mlx5_hca_attr {
 	uint32_t regex:1;
 	uint32_t regexp_num_of_engines;
 	uint32_t log_max_ft_sampler_num:8;
+	uint32_t geneve_tlv_opt;
 	struct mlx5_hca_qos_attr qos;
 	struct mlx5_hca_vdpa_attr vdpa;
 	int log_max_qp_sz;
@@ -479,6 +482,7 @@ struct mlx5_devx_obj *mlx5_devx_cmd_create_flex_parser(void *ctx,
 __rte_internal
 int mlx5_devx_cmd_register_read(void *ctx, uint16_t reg_id,
 				uint32_t arg, uint32_t *data, uint32_t dw_cnt);
+
 /**
  * Create virtio queue counters object DevX API.
  *
drivers/common/mlx5/mlx5_prm.h +25 -3
index 8c9b53ce10..458550279e 100644
@@ -789,7 +789,7 @@ struct mlx5_ifc_fte_match_set_misc3_bits {
 	u8 icmp_code[0x8];
 	u8 icmpv6_type[0x8];
 	u8 icmpv6_code[0x8];
-	u8 reserved_at_120[0x20];
+	u8 geneve_tlv_option_0_data[0x20];
 	u8 gtpu_teid[0x20];
 	u8 gtpu_msg_type[0x08];
 	u8 gtpu_msg_flags[0x08];
@@ -1082,6 +1082,8 @@ enum {
 			(1ULL << MLX5_GENERAL_OBJ_TYPE_FLEX_PARSE_GRAPH)
 #define MLX5_GENERAL_OBJ_TYPES_CAP_FLOW_HIT_ASO \
 			(1ULL << MLX5_GENERAL_OBJ_TYPE_FLOW_HIT_ASO)
+#define MLX5_GENERAL_OBJ_TYPES_CAP_GENEVE_TLV_OPT \
+			(1ULL << MLX5_OBJ_TYPE_GENEVE_TLV_OPT)
 
 enum {
 	MLX5_HCA_CAP_OPMOD_GET_MAX   = 0,
@@ -1380,8 +1382,10 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 	u8 reserved_at_500[0x20];
 	u8 num_of_uars_per_page[0x20];
 	u8 flex_parser_protocols[0x20];
-	u8 reserved_at_560[0x20];
-	u8 reserved_at_580[0x3c];
+	u8 max_geneve_tlv_options[0x8];
+	u8 reserved_at_568[0x3];
+	u8 max_geneve_tlv_option_data_len[0x5];
+	u8 reserved_at_570[0x4c];
 	u8 mini_cqe_resp_stride_index[0x1];
 	u8 cqe_128_always[0x1];
 	u8 cqe_compression_128[0x1];
@@ -2292,6 +2296,7 @@ struct mlx5_ifc_create_cq_in_bits {
 };
 
 enum {
+	MLX5_OBJ_TYPE_GENEVE_TLV_OPT = 0x000b,
 	MLX5_GENERAL_OBJ_TYPE_VIRTQ = 0x000d,
 	MLX5_GENERAL_OBJ_TYPE_VIRTIO_Q_COUNTERS = 0x001c,
 	MLX5_GENERAL_OBJ_TYPE_FLEX_PARSE_GRAPH = 0x0022,
@@ -2326,6 +2331,17 @@ struct mlx5_ifc_virtio_q_counters_bits {
 	u8 reserved_at_180[0x50];
 };
 
+struct mlx5_ifc_geneve_tlv_option_bits {
+	u8 modify_field_select[0x40];
+	u8 reserved_at_40[0x18];
+	u8 geneve_option_fte_index[0x8];
+	u8 option_class[0x10];
+	u8 option_type[0x8];
+	u8 reserved_at_78[0x3];
+	u8 option_data_length[0x5];
+	u8 reserved_at_80[0x180];
+};
+
 struct mlx5_ifc_create_virtio_q_counters_in_bits {
 	struct mlx5_ifc_general_obj_in_cmd_hdr_bits hdr;
 	struct mlx5_ifc_virtio_q_counters_bits virtio_q_counters;
@@ -2335,6 +2351,12 @@ struct mlx5_ifc_query_virtio_q_counters_out_bits {
 	struct mlx5_ifc_general_obj_in_cmd_hdr_bits hdr;
 	struct mlx5_ifc_virtio_q_counters_bits virtio_q_counters;
 };
+
+struct mlx5_ifc_create_geneve_tlv_option_in_bits {
+	struct mlx5_ifc_general_obj_in_cmd_hdr_bits hdr;
+	struct mlx5_ifc_geneve_tlv_option_bits geneve_tlv_opt;
+};
+
 enum {
 	MLX5_VIRTQ_STATE_INIT = 0,
 	MLX5_VIRTQ_STATE_RDY = 1,
Signature · 1 line
2.21.0
#8 lore raw

PATCH v3 4/8 common/mlx5: create GENEVE TLV option object with DevX

In reply to Shiri Kuzin

TLV object is a special firmware maintained entity used to support match on GENEVE header extension option.

The TLV object is created with DevX API and accepts the option class, type and lehgth fields.

The class type and length fields are set using MLX5_SET and the Devx object is created using mlx5 glue function.

---
 drivers/common/mlx5/mlx5_devx_cmds.c | 56 +++++++++++++++++++++++++++-
 drivers/common/mlx5/mlx5_devx_cmds.h |  5 +++
 drivers/common/mlx5/version.map      |  1 +
 3 files changed, 61 insertions(+), 1 deletion(-)
drivers/common/mlx5/mlx5_devx_cmds.c +55 -1
index fc406af062..08bd357ad1 100644
@@ -2068,7 +2068,6 @@ mlx5_devx_cmd_create_flow_hit_aso_obj(void *ctx, uint32_t pd)
  *
  * @param[in] ctx
  *   Context returned from mlx5 open_device() glue function.
- *
  * @return
  *   The DevX object created, NULL otherwise and rte_errno is set.
  */
@@ -2097,3 +2096,58 @@ mlx5_devx_cmd_alloc_pd(void *ctx)
 	ppd->id = MLX5_GET(alloc_pd_out, out, pd);
 	return ppd;
 }
+
+/**
+ * Create general object of type GENEVE TLV option using DevX API.
+ *
+ * @param[in] ctx
+ *   Context returned from mlx5 open_device() glue function.
+ * @param [in] class
+ *   TLV option variable value of class
+ * @param [in] type
+ *   TLV option variable value of type
+ * @param [in] len
+ *   TLV option variable value of len
+ *    *
+ * @return
+ *   The DevX object created, NULL otherwise and rte_errno is set.
+ */
+struct mlx5_devx_obj *
+mlx5_devx_cmd_create_geneve_tlv_option(void *ctx,
+		uint16_t class, uint8_t type, uint8_t len)
+{
+	uint32_t in[MLX5_ST_SZ_DW(create_geneve_tlv_option_in)] = {0};
+	uint32_t out[MLX5_ST_SZ_DW(general_obj_out_cmd_hdr)] = {0};
+	struct mlx5_devx_obj *geneve_tlv_opt_obj = mlx5_malloc(MLX5_MEM_ZERO,
+						   sizeof(*geneve_tlv_opt_obj),
+						   0, SOCKET_ID_ANY);
+
+	if (!geneve_tlv_opt_obj) {
+		DRV_LOG(ERR, "Failed to allocate geneve tlv option object.");
+		rte_errno = ENOMEM;
+		return NULL;
+	}
+	void *hdr = MLX5_ADDR_OF(create_geneve_tlv_option_in, in, hdr);
+	void *opt = MLX5_ADDR_OF(create_geneve_tlv_option_in, in,
+			geneve_tlv_opt);
+	MLX5_SET(general_obj_in_cmd_hdr, hdr, opcode,
+			MLX5_CMD_OP_CREATE_GENERAL_OBJECT);
+	MLX5_SET(general_obj_in_cmd_hdr, hdr, obj_type,
+			MLX5_OBJ_TYPE_GENEVE_TLV_OPT);
+	MLX5_SET(geneve_tlv_option, opt, option_class,
+			rte_be_to_cpu_16(class));
+	MLX5_SET(geneve_tlv_option, opt, option_type, type);
+	MLX5_SET(geneve_tlv_option, opt, option_data_length, len);
+	geneve_tlv_opt_obj->obj = mlx5_glue->devx_obj_create(ctx, in,
+					sizeof(in), out, sizeof(out));
+	if (!geneve_tlv_opt_obj->obj) {
+		rte_errno = errno;
+		DRV_LOG(ERR, "Failed to create Geneve tlv option "
+				"Obj using DevX.");
+		mlx5_free(geneve_tlv_opt_obj);
+		return NULL;
+	}
+	geneve_tlv_opt_obj->id = MLX5_GET(general_obj_out_cmd_hdr, out, obj_id);
+	return geneve_tlv_opt_obj;
+}
+
drivers/common/mlx5/mlx5_devx_cmds.h +5 -0
index 6b2f77837b..6774b49d2d 100644
@@ -483,6 +483,11 @@ __rte_internal
 int mlx5_devx_cmd_register_read(void *ctx, uint16_t reg_id,
 				uint32_t arg, uint32_t *data, uint32_t dw_cnt);
 
+__rte_internal
+struct mlx5_devx_obj *
+mlx5_devx_cmd_create_geneve_tlv_option(void *ctx,
+		uint16_t class, uint8_t type, uint8_t len);
+
 /**
  * Create virtio queue counters object DevX API.
  *
drivers/common/mlx5/version.map +1 -0
index fb3952bbc6..7de8006b89 100644
@@ -23,6 +23,7 @@ INTERNAL {
 	mlx5_devx_cmd_create_virtio_q_counters;
 	mlx5_devx_cmd_create_virtq;
         mlx5_devx_cmd_create_flow_hit_aso_obj;
+	mlx5_devx_cmd_create_geneve_tlv_option;
 	mlx5_devx_cmd_destroy;
 	mlx5_devx_cmd_flow_counter_alloc;
 	mlx5_devx_cmd_flow_counter_query;
Signature · 1 line
2.21.0
#9 lore raw

PATCH v3 5/8 net/mlx5: create GENEVE TLV option management

In reply to Shiri Kuzin

Currently firmware supports the only TLV object per device to match on the GENEVE header option.

This patch adds the simple TLV object management to the mlx5 PMD.

---
 drivers/net/mlx5/mlx5.c         |   2 +
 drivers/net/mlx5/mlx5.h         |  13 ++++
 drivers/net/mlx5/mlx5_flow.h    |   4 ++
 drivers/net/mlx5/mlx5_flow_dv.c | 108 ++++++++++++++++++++++++++++++++
 4 files changed, 127 insertions(+)
drivers/net/mlx5/mlx5.c +2 -0
index 023ef50a77..68d6352d48 100644
@@ -1014,6 +1014,7 @@ mlx5_alloc_shared_dev_ctx(const struct mlx5_dev_spawn_data *spawn,
 	rte_rwlock_write_unlock(&mlx5_shared_data->mem_event_rwlock);
 	/* Add context to the global device list. */
 	LIST_INSERT_HEAD(&mlx5_dev_ctx_list, sh, next);
+	rte_spinlock_init(&sh->geneve_tlv_opt_sl);
 exit:
 	pthread_mutex_unlock(&mlx5_dev_ctx_list_mutex);
 	return sh;
@@ -1109,6 +1110,7 @@ mlx5_free_shared_dev_ctx(struct mlx5_dev_ctx_shared *sh)
 		mlx5_glue->devx_free_uar(sh->devx_rx_uar);
 	if (sh->ctx)
 		claim_zero(mlx5_glue->close_device(sh->ctx));
+	MLX5_ASSERT(sh->geneve_tlv_option_resource == NULL);
 	pthread_mutex_destroy(&sh->txpp.mutex);
 	mlx5_free(sh);
 	return;
drivers/net/mlx5/mlx5.h +13 -0
index 41034f5d19..23272950a4 100644
@@ -540,6 +540,16 @@ struct mlx5_aso_age_mng {
 	struct mlx5_aso_sq aso_sq; /* ASO queue objects. */
 };
 
+/* Management structure for geneve tlv option */
+struct mlx5_geneve_tlv_option_resource {
+	struct mlx5_devx_obj *obj; /* Pointer to the geneve tlv opt object. */
+	rte_be16_t option_class; /* geneve tlv opt class.*/
+	uint8_t option_type; /* geneve tlv opt type.*/
+	uint8_t length; /* geneve tlv opt length. */
+	uint32_t refcnt; /* geneve tlv object reference counter */
+};
+
+
 #define MLX5_AGE_EVENT_NEW		1
 #define MLX5_AGE_TRIGGER		2
 #define MLX5_AGE_SET(age_info, BIT) \
@@ -752,6 +762,9 @@ struct mlx5_dev_ctx_shared {
 	void *devx_rx_uar; /* DevX UAR for Rx. */
 	struct mlx5_aso_age_mng *aso_age_mng;
 	/* Management data for aging mechanism using ASO Flow Hit. */
+	struct mlx5_geneve_tlv_option_resource *geneve_tlv_option_resource;
+	/* Management structure for geneve tlv option */
+	rte_spinlock_t geneve_tlv_opt_sl; /* Lock for geneve tlv resource */
 	struct mlx5_dev_shared_port port[]; /* per device port data array. */
 };
 
drivers/net/mlx5/mlx5_flow.h +4 -0
index ee85c9d8a5..0c8861964c 100644
@@ -1046,6 +1046,7 @@ struct rte_flow {
 	uint32_t counter; /**< Holds flow counter. */
 	uint32_t tunnel_id;  /**< Tunnel id */
 	uint32_t age; /**< Holds ASO age bit index. */
+	uint32_t geneve_tlv_option; /**< Holds Geneve TLV option id. > */
 } __rte_packed;
 
 /*
@@ -1503,4 +1504,7 @@ void flow_dv_dest_array_remove_cb(struct mlx5_cache_list *list,
 				  struct mlx5_cache_entry *entry);
 struct mlx5_aso_age_action *flow_aso_age_get_by_idx(struct rte_eth_dev *dev,
 						    uint32_t age_idx);
+int flow_dev_geneve_tlv_option_resource_register(struct rte_eth_dev *dev,
+					     const struct rte_flow_item *item,
+					     struct rte_flow_error *error);
 #endif /* RTE_PMD_MLX5_FLOW_H_ */
drivers/net/mlx5/mlx5_flow_dv.c +108 -0

126 lines not on this page · raw message

Signature · 1 line
2.21.0
#10 lore raw

PATCH v3 6/8 net/mlx5: add GENEVE TLV option flow validation

In reply to Shiri Kuzin

This patch adds validation routine for the GENEVE header TLV option.

The GENEVE TLV option match must include all fields with full masks due to NIC does not support masking on option class, type and length.

The option data length must be non zero and provided data pattern should be zero neither due to hardware limitations.

---
 drivers/net/mlx5/mlx5_flow.c    | 120 ++++++++++++++++++++++++++++++++
 drivers/net/mlx5/mlx5_flow.h    |   7 ++
 drivers/net/mlx5/mlx5_flow_dv.c |  10 ++-
 3 files changed, 136 insertions(+), 1 deletion(-)
drivers/net/mlx5/mlx5_flow.c +120 -0
index 2a4073c126..892ed2f2d9 100644
@@ -2627,6 +2627,126 @@ mlx5_flow_validate_item_geneve(const struct rte_flow_item *item,
 	return 0;
 }
 
+/**
+ * Validate Geneve TLV option item.
+ *
+ * @param[in] item
+ *   Item specification.
+ * @param[in] last_item
+ *   Previous validated item in the pattern items.
+ * @param[in] dev
+ *   Pointer to the rte_eth_dev structure.
+ * @param[out] error
+ *   Pointer to error structure.
+ *
+ * @return
+ *   0 on success, a negative errno value otherwise and rte_errno is set.
+ */
+int
+mlx5_flow_validate_item_geneve_opt(const struct rte_flow_item *item,
+				   uint64_t last_item,
+				   struct rte_eth_dev *dev,
+				   struct rte_flow_error *error)
+{
+	struct mlx5_priv *priv = dev->data->dev_private;
+	struct mlx5_dev_ctx_shared *sh = priv->sh;
+	struct mlx5_geneve_tlv_option_resource *geneve_opt_resource;
+	struct mlx5_hca_attr *hca_attr = &priv->config.hca_attr;
+	uint8_t data_max_supported =
+			hca_attr->max_geneve_tlv_option_data_len * 4;
+	struct mlx5_dev_config *config = &priv->config;
+	const struct rte_flow_item_geneve_opt *spec = item->spec;
+	const struct rte_flow_item_geneve_opt *mask = item->mask;
+	unsigned int i;
+	unsigned int data_len;
+	const struct rte_flow_item_geneve_opt full_mask = {
+		.option_class = RTE_BE16(0xffff),
+		.option_type = 0xff,
+		.option_len = 0x1f,
+	};
+
+	if (!mask)
+		mask = &rte_flow_item_geneve_opt_mask;
+	if (!spec)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt class/type/length must be specified");
+	if ((uint32_t)(spec->option_len) > MLX5_GENEVE_OPTLEN_MASK)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt length exceeeds the limit (31)");
+	/* Check if class type and length masks are full. */
+	if (full_mask.option_class != mask->option_class ||
+	    full_mask.option_type != mask->option_type ||
+	    full_mask.option_len != (mask->option_len & full_mask.option_len))
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt class/type/length masks must be full");
+	/* Check if length is supported */
+	if ((uint32_t)(spec->option_len) >
+			config->hca_attr.max_geneve_tlv_option_data_len)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt length not supported");
+	if (config->hca_attr.max_geneve_tlv_options > 1)
+		DRV_LOG(DEBUG,
+			"max_geneve_tlv_options supports more than 1 option");
+	/* Check GENEVE item preceding. */
+	if (!(last_item & MLX5_FLOW_LAYER_GENEVE))
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve opt item must be preceded with Geneve item");
+	/* Check if length is 0 or data is 0. */
+	if (spec->data == NULL || spec->option_len == 0)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt with zero data/length not supported");
+	/* Check not all data & mask are 0. */
+	data_len = spec->option_len * 4;
+	if (mask->data == NULL) {
+		for (i = 0; i < data_len; i++)
+			if (spec->data[i])
+				break;
+		if (i == data_len)
+			return rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_ITEM, item,
+				"Can't match on Geneve option data 0");
+	} else {
+		for (i = 0; i < data_len; i++)
+			if (spec->data[i] & mask->data[i])
+				break;
+		if (i == data_len)
+			return rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_ITEM, item,
+				"Can't match on Geneve option data and mask 0");
+		/* Check data mask supported. */
+		for (i = data_max_supported; i < data_len ; i++)
+			if (mask->data[i])
+				return rte_flow_error_set(error, ENOTSUP,
+					RTE_FLOW_ERROR_TYPE_ITEM, item,
+					"Data mask is of unsupported size");
+	}
+	/* Check GENEVE option is supported in NIC. */
+	if (!config->hca_attr.geneve_tlv_opt)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt not supported");
+	/* Check if we already have geneve option with different type/class. */
+	rte_spinlock_lock(&sh->geneve_tlv_opt_sl);
+	geneve_opt_resource = sh->geneve_tlv_option_resource;
+	if (geneve_opt_resource != NULL)
+		if (geneve_opt_resource->option_class != spec->option_class ||
+		    geneve_opt_resource->option_type != spec->option_type ||
+		    geneve_opt_resource->length != spec->option_len) {
+			rte_spinlock_unlock(&sh->geneve_tlv_opt_sl);
+			return rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_ITEM, item,
+				"Only one Geneve TLV option supported");
+		}
+	rte_spinlock_unlock(&sh->geneve_tlv_opt_sl);
+	return 0;
+}
+
 /**
  * Validate MPLS item.
  *
drivers/net/mlx5/mlx5_flow.h +7 -0
index 0c8861964c..28db5faee1 100644
@@ -138,6 +138,9 @@ enum mlx5_feature_name {
 #define MLX5_FLOW_LAYER_OUTER_L3_IPV6_FRAG_EXT (1u << 30)
 #define MLX5_FLOW_LAYER_INNER_L3_IPV6_FRAG_EXT (1u << 31)
 
+/* Pattern tunnel Layer bits (continued). */
+#define MLX5_FLOW_LAYER_GENEVE_OPT (UINT64_C(1) << 32)
+
 /* Outer Masks. */
 #define MLX5_FLOW_LAYER_OUTER_L3 \
 	(MLX5_FLOW_LAYER_OUTER_L3_IPV4 | MLX5_FLOW_LAYER_OUTER_L3_IPV6)
@@ -1396,6 +1399,10 @@ int mlx5_flow_validate_item_geneve(const struct rte_flow_item *item,
 				   uint64_t item_flags,
 				   struct rte_eth_dev *dev,
 				   struct rte_flow_error *error);
+int mlx5_flow_validate_item_geneve_opt(const struct rte_flow_item *item,
+				   uint64_t last_item,
+				   struct rte_eth_dev *dev,
+				   struct rte_flow_error *error);
 int mlx5_flow_validate_item_ecpri(const struct rte_flow_item *item,
 				  uint64_t item_flags,
 				  uint64_t last_item,
drivers/net/mlx5/mlx5_flow_dv.c +9 -1
index 0a657cea41..c24123fbf3 100644
@@ -5520,6 +5520,15 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr,
 				return ret;
 			last_item = MLX5_FLOW_LAYER_GENEVE;
 			break;
+		case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+			ret = mlx5_flow_validate_item_geneve_opt(items,
+								 last_item,
+								 dev,
+								 error);
+			if (ret < 0)
+				return ret;
+			last_item = MLX5_FLOW_LAYER_GENEVE_OPT;
+			break;
 		case RTE_FLOW_ITEM_TYPE_MPLS:
 			ret = mlx5_flow_validate_item_mpls(dev, items,
 							   item_flags,
@@ -5528,7 +5537,6 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr,
 				return ret;
 			last_item = MLX5_FLOW_LAYER_MPLS;
 			break;
-
 		case RTE_FLOW_ITEM_TYPE_MARK:
 			ret = flow_dv_validate_item_mark(dev, items, attr,
 							 error);
Signature · 1 line
2.21.0
#11 lore raw

PATCH v3 7/8 net/mlx5: add GENEVE TLV option flow translation

In reply to Shiri Kuzin

The GENEVE TLV option matching flows must be created using a translation function.

This function checks whether we already created a Devx object for the matching and either creates the objects or updates the reference counter.

---
 drivers/net/mlx5/mlx5_flow_dv.c | 70 +++++++++++++++++++++++++++++++++
 1 file changed, 70 insertions(+)
drivers/net/mlx5/mlx5_flow_dv.c +70 -0
index c24123fbf3..006460cd0a 100644
@@ -7336,6 +7336,65 @@ flow_dev_geneve_tlv_option_resource_register(struct rte_eth_dev *dev,
 	return ret;
 }
 
+/**
+ * Add Geneve TLV option item to matcher.
+ *
+ * @param[in, out] dev
+ *   Pointer to rte_eth_dev structure.
+ * @param[in, out] matcher
+ *   Flow matcher.
+ * @param[in, out] key
+ *   Flow matcher value.
+ * @param[in] item
+ *   Flow pattern to translate.
+ * @param[out] error
+ *   Pointer to error structure.
+ */
+static int
+flow_dv_translate_item_geneve_opt(struct rte_eth_dev *dev, void *matcher,
+				  void *key, const struct rte_flow_item *item,
+				  struct rte_flow_error *error)
+{
+	const struct rte_flow_item_geneve_opt *geneve_opt_m = item->mask;
+	const struct rte_flow_item_geneve_opt *geneve_opt_v = item->spec;
+	void *misc3_m = MLX5_ADDR_OF(fte_match_param, matcher,
+			misc_parameters_3);
+	void *misc3_v = MLX5_ADDR_OF(fte_match_param, key, misc_parameters_3);
+	rte_be32_t opt_data_key = 0, opt_data_mask = 0;
+	int ret = 0;
+
+	if (!geneve_opt_v)
+		return -1;
+	if (!geneve_opt_m)
+		geneve_opt_m = &rte_flow_item_geneve_opt_mask;
+	ret = flow_dev_geneve_tlv_option_resource_register(dev, item,
+							   error);
+	if (ret) {
+		DRV_LOG(ERR, "Failed to create geneve_tlv_obj");
+		return ret;
+	}
+	/* Set the data. */
+	if (geneve_opt_v->data) {
+		memcpy(&opt_data_key, geneve_opt_v->data,
+			RTE_MIN((uint32_t)(geneve_opt_v->option_len * 4),
+				sizeof(opt_data_key)));
+		MLX5_ASSERT((uint32_t)(geneve_opt_v->option_len * 4) <=
+				sizeof(opt_data_key));
+		memcpy(&opt_data_mask, geneve_opt_m->data,
+			RTE_MIN((uint32_t)(geneve_opt_v->option_len * 4),
+				sizeof(opt_data_mask)));
+		MLX5_ASSERT((uint32_t)(geneve_opt_v->option_len * 4) <=
+				sizeof(opt_data_mask));
+		MLX5_SET(fte_match_set_misc3, misc3_m,
+				geneve_tlv_option_0_data,
+				rte_be_to_cpu_32(opt_data_mask));
+		MLX5_SET(fte_match_set_misc3, misc3_v,
+				geneve_tlv_option_0_data,
+			rte_be_to_cpu_32(opt_data_key & opt_data_mask));
+	}
+	return ret;
+}
+
 /**
  * Add MPLS item to matcher and to the value.
  *
@@ -10583,6 +10642,17 @@ flow_dv_translate(struct rte_eth_dev *dev,
 			matcher.priority = MLX5_TUNNEL_PRIO_GET(rss_desc);
 			last_item = MLX5_FLOW_LAYER_GENEVE;
 			break;
+		case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+			ret = flow_dv_translate_item_geneve_opt(dev, match_mask,
+							  match_value,
+							  items, error);
+			if (ret)
+				return rte_flow_error_set(error, -ret,
+					RTE_FLOW_ERROR_TYPE_ITEM, NULL,
+					"cannot create GENEVE TLV option");
+			flow->geneve_tlv_option = 1;
+			last_item = MLX5_FLOW_LAYER_GENEVE_OPT;
+			break;
 		case RTE_FLOW_ITEM_TYPE_MPLS:
 			flow_dv_translate_item_mpls(match_mask, match_value,
 						    items, last_item, tunnel);
Signature · 1 line
2.21.0
#12 lore raw

PATCH v3 8/8 doc: update GENEVE TLV option support

In reply to Shiri Kuzin

GENEVE TLV option support added to mlx5 PMD.

The limitations and support were updated in
documentation.

---
 doc/guides/nics/mlx5.rst               | 23 ++++++++++++++++++++++-
 doc/guides/rel_notes/release_21_02.rst |  8 ++++++++
 2 files changed, 30 insertions(+), 1 deletion(-)
doc/guides/nics/mlx5.rst +22 -1
index 3bda0f8417..5f2b42b979 100644
@@ -98,6 +98,7 @@ Features
 - Hardware LRO.
 - Hairpin.
 - Multiple-thread flow insertion.
+- Matching on Geneve TLV option header with raw encap/decap action.
 
 Limitations
 -----------
@@ -175,7 +176,18 @@ Limitations
      - OAM
      - protocol type
      - options length
-       Currently, the only supported options length value is 0.
+
+- Match on Geneve TLv option is supported on the following fields:
+     - Class
+     - Type
+     - Length
+     - Data
+
+  Only one Class/Type/Length Geneve TLV option is supported per shared device.
+  Class/Type/Length fields must be specified as well as masks.
+  Class/Type/Length specified masks must be full.
+  Matching Geneve TLV option without specifying data is not supported.
+  Matching Geneve TLV option with data & mask == 0 is not supported.
 
 - VF: flow rules created on VF devices can only match traffic targeted at the
   configured MAC addresses (see ``rte_eth_dev_mac_addr_add()``).
@@ -1022,6 +1034,10 @@ Below are some firmware configurations listed.
    or
    FLEX_PARSER_PROFILE_ENABLE=1
 
+- enable Geneve TLV option flow matching::
+
+   FLEX_PARSER_PROFILE_ENABLE=0
+
 - enable GTP flow matching::
 
    FLEX_PARSER_PROFILE_ENABLE=3
@@ -1501,6 +1517,11 @@ Supported hardware offloads
    |                       | |  rdma-core 32 | | rdma-core 32  |
    |                       | |  ConnectX-6 Dx| | ConnectX-6 Dx |
    +-----------------------+-----------------+-----------------+
+   | Encapsulation         | | DPDK 21.02    | | DPDK 21.02    |
+   | GENEVE TLV option     | | OFED 5.2      | | OFED 5.2      |
+   |                       | | rdma-core 34  | | rdma-core 34  |
+   |                       | | ConnectX-6 Dx | | ConnectX-6 Dx |
+   +-----------------------+-----------------+-----------------+
 
 Notes for metadata
 ------------------
doc/guides/rel_notes/release_21_02.rst +8 -0
index 638f98168b..2fb5bf9c38 100644
@@ -55,6 +55,14 @@ New Features
      Also, make sure to start the actual text at the margin.
      =======================================================
 
+* **Updated Mellanox mlx5 driver.**
+
+  Updated the Mellanox mlx5 driver with new features and improvements, including:
+
+  * **Added GENEVE TLV option in rte_flow.**
+
+  Added support for matching GENEVE TLV option and raw encap/decap of GENEVE
+  TLV option.
 
 Removed Items
 -------------
Signature · 1 line
2.21.0
#13 lore raw

v4 cover ethdev: introduce GENEVE header TLV option item

In reply to Shiri Kuzin

The Geneve tunneling protocol is designed to allow the user to specify some data context on the packet.
The GENEVE TLV (Type-Length-Variable) Option is the mean intended to present the user data.

In order to support GENEVE TLV Option the new rte_flow item "rte_flow_item_geneve_opt" is introduced.
The new item contains the values and masks for the following fields:
-option class
-option type
-length
-data

The usage example:
"flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 / geneve-opt class is 5 length is 1 type is 0 data is 0x66998800 / end actions count / drop / end"

New item will be added to testpmd to support raw encap/decap action.

v4:
- fixed setting option length.
- fixed testpmd calculation of GENEVE option size.
- updated documentation.

v3:
- updated documentation.

v2:
- removed pedantic.

Shiri Kuzin (7):
lib/librte_ethdev: introduce GENEVE header TLV option item common/mlx5: check GENEVE TLV support in HCA attributes common/mlx5: create GENEVE TLV option object with DevX net/mlx5: create GENEVE TLV option management net/mlx5: add GENEVE TLV option flow validation net/mlx5: add GENEVE TLV option flow translation
doc: update GENEVE TLV option support

Viacheslav Ovsiienko (1):
app/testpmd: add GENEVE option item support

 app/test-pmd/cmdline_flow.c                 | 101 +++++++++-
 doc/guides/nics/mlx5.rst                    |  23 ++-
 doc/guides/rel_notes/release_21_02.rst      |   8 +
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |   8 +
 drivers/common/mlx5/mlx5_devx_cmds.c        |  63 ++++++-
 drivers/common/mlx5/mlx5_devx_cmds.h        |   9 +
 drivers/common/mlx5/mlx5_prm.h              |  28 ++-
 drivers/common/mlx5/version.map             |   1 +
 drivers/net/mlx5/mlx5.c                     |   2 +
 drivers/net/mlx5/mlx5.h                     |  13 ++
 drivers/net/mlx5/mlx5_flow.c                | 120 ++++++++++++
 drivers/net/mlx5/mlx5_flow.h                |  13 +-
 drivers/net/mlx5/mlx5_flow_dv.c             | 198 +++++++++++++++++++-
 lib/librte_ethdev/rte_flow.c                |   1 +
 lib/librte_ethdev/rte_flow.h                |  27 +++
 15 files changed, 602 insertions(+), 13 deletions(-)
Signature · 1 line
2.21.0
#14 lore raw

PATCH v4 1/8 lib/librte_ethdev: introduce GENEVE header TLV option item

The Geneve tunneling protocol is designed to allow the user to specify some data context on the packet. The GENEVE TLV (Type-Length-Variable) Option is the mean intended to present the user data.

In order to support GENEVE TLV Option the new rte_flow item "rte_flow_item_geneve_opt" is added.
The new item contains the values and masks for the following fields:
-option class
-option type
-length
-data

New item will be added to testpmd to support match and raw encap/decap actions.

---
 lib/librte_ethdev/rte_flow.c |  1 +
 lib/librte_ethdev/rte_flow.h | 27 +++++++++++++++++++++++++++
 2 files changed, 28 insertions(+)
lib/librte_ethdev/rte_flow.c +1 -0
index a06f64c271..2af7d965e1 100644
@@ -97,6 +97,7 @@ static const struct rte_flow_desc_data rte_flow_desc_item[] = {
 	MK_FLOW_ITEM(L2TPV3OIP, sizeof(struct rte_flow_item_l2tpv3oip)),
 	MK_FLOW_ITEM(PFCP, sizeof(struct rte_flow_item_pfcp)),
 	MK_FLOW_ITEM(ECPRI, sizeof(struct rte_flow_item_ecpri)),
+	MK_FLOW_ITEM(GENEVE_OPT, sizeof(struct rte_flow_item_geneve_opt)),
 };
 
 /** Generate flow_action[] entry. */
lib/librte_ethdev/rte_flow.h +27 -0
index 0977a78270..11a6494b8e 100644
@@ -543,6 +543,14 @@ enum rte_flow_item_type {
 	 * See struct rte_flow_item_ipv6_frag_ext.
 	 */
 	RTE_FLOW_ITEM_TYPE_IPV6_FRAG_EXT,
+
+	/**
+	 * Matches Geneve Variable Length Option
+	 *
+	 * See struct rte_flow_item_geneve_opt
+	 */
+	RTE_FLOW_ITEM_TYPE_GENEVE_OPT,
+
 };
 
 /**
@@ -1627,6 +1635,25 @@ static const struct rte_flow_item_ecpri rte_flow_item_ecpri_mask = {
 };
 #endif
 
+/**
+ * RTE_FLOW_ITEM_TYPE_GENEVE_OPT
+ *
+ * Matches a GENEVE Variable Length Option
+ */
+struct rte_flow_item_geneve_opt {
+	rte_be16_t option_class;
+	uint8_t option_type;
+	uint8_t option_len;
+	uint32_t *data;
+};
+
+/** Default mask for RTE_FLOW_ITEM_TYPE_GENEVE_OPT. */
+#ifndef __cplusplus
+static const struct rte_flow_item_geneve_opt
+rte_flow_item_geneve_opt_mask = {
+	.option_type = 0xff,
+};
+#endif
 /**
  * Matching pattern item definition.
  *
Signature · 1 line
2.21.0
#15 lore raw

PATCH v4 2/8 app/testpmd: add GENEVE option item support

In reply to Shiri Kuzin

From: Viacheslav Ovsiienko <[email protected]>

The patch adds the GENEVE option rte flow item support to command line interpreter. The flow command with GENEVE option items looks like:

flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 /
geneve-opt class is 99 length is 1 type is 0 data is 0x669988 / end actions drop / end

The option length should be specified in 32-bit words, this value specifies the length of the data pattern/mask arrays (should be multiplied by sizeof(uint32_t) to be expressed in bytes. If match on the length itself is not needed the mask should be set to zero, in this case length is used to specify the pattern/mask array lengths only.

---
 app/test-pmd/cmdline_flow.c                 | 101 ++++++++++++++++++--
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |   8 ++
 2 files changed, 103 insertions(+), 6 deletions(-)
app/test-pmd/cmdline_flow.c +95 -6
index 585cab98b4..98afde82cc 100644
@@ -283,6 +283,11 @@ enum index {
 	ITEM_ECPRI_MSG_IQ_DATA_PCID,
 	ITEM_ECPRI_MSG_RTC_CTRL_RTCID,
 	ITEM_ECPRI_MSG_DLY_MSR_MSRID,
+	ITEM_GENEVE_OPT,
+	ITEM_GENEVE_OPT_CLASS,
+	ITEM_GENEVE_OPT_TYPE,
+	ITEM_GENEVE_OPT_LENGTH,
+	ITEM_GENEVE_OPT_DATA,
 
 	/* Validate/create actions. */
 	ACTIONS,
@@ -413,6 +418,9 @@ enum index {
 /** Maximum size for pattern in struct rte_flow_item_raw. */
 #define ITEM_RAW_PATTERN_SIZE 40
 
+/** Maximum size for GENEVE option data pattern in bytes. */
+#define ITEM_GENEVE_OPT_DATA_SIZE 124
+
 /** Storage size for struct rte_flow_item_raw including pattern. */
 #define ITEM_RAW_SIZE \
 	(sizeof(struct rte_flow_item_raw) + ITEM_RAW_PATTERN_SIZE)
@@ -428,7 +436,7 @@ struct action_rss_data {
 };
 
 /** Maximum data size in struct rte_flow_action_raw_encap. */
-#define ACTION_RAW_ENCAP_MAX_DATA 128
+#define ACTION_RAW_ENCAP_MAX_DATA 512
 #define RAW_ENCAP_CONFS_MAX_NUM 8
 
 /** Storage for struct rte_flow_action_raw_encap. */
@@ -658,6 +666,16 @@ struct token {
 		.mask = (const void *)&(const s){ .f = (1 << (b)) - 1 }, \
 	})
 
+/** Static initializer for ARGS() to target a field with limits. */
+#define ARGS_ENTRY_BOUNDED(s, f, i, a) \
+	(&(const struct arg){ \
+		.bounded = 1, \
+		.min = (i), \
+		.max = (a), \
+		.offset = offsetof(s, f), \
+		.size = sizeof(((s *)0)->f), \
+	})
+
 /** Static initializer for ARGS() to target an arbitrary bit-mask. */
 #define ARGS_ENTRY_MASK(s, f, m) \
 	(&(const struct arg){ \
@@ -903,6 +921,7 @@ static const enum index next_item[] = {
 	ITEM_AH,
 	ITEM_PFCP,
 	ITEM_ECPRI,
+	ITEM_GENEVE_OPT,
 	END_SET,
 	ZERO,
 };
@@ -1244,6 +1263,15 @@ static const enum index item_ecpri_common_type[] = {
 	ZERO,
 };
 
+static const enum index item_geneve_opt[] = {
+	ITEM_GENEVE_OPT_CLASS,
+	ITEM_GENEVE_OPT_TYPE,
+	ITEM_GENEVE_OPT_LENGTH,
+	ITEM_GENEVE_OPT_DATA,
+	ITEM_NEXT,
+	ZERO,
+};
+
 static const enum index next_action[] = {
 	ACTION_END,
 	ACTION_VOID,
@@ -3230,6 +3258,47 @@ static const struct token token_list[] = {
 		.args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_ecpri,
 				hdr.type5.msr_id)),
 	},
+	[ITEM_GENEVE_OPT] = {
+		.name = "geneve-opt",
+		.help = "GENEVE header option",
+		.priv = PRIV_ITEM(GENEVE_OPT,
+				  sizeof(struct rte_flow_item_geneve_opt) +
+				  ITEM_GENEVE_OPT_DATA_SIZE),
+		.next = NEXT(item_geneve_opt),
+		.call = parse_vc,
+	},
+	[ITEM_GENEVE_OPT_CLASS]	= {
+		.name = "class",
+		.help = "GENEVE option class",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(UNSIGNED), item_param),
+		.args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_geneve_opt,
+					     option_class)),
+	},
+	[ITEM_GENEVE_OPT_TYPE] = {
+		.name = "type",
+		.help = "GENEVE option type",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(UNSIGNED), item_param),
+		.args = ARGS(ARGS_ENTRY(struct rte_flow_item_geneve_opt,
+					option_type)),
+	},
+	[ITEM_GENEVE_OPT_LENGTH] = {
+		.name = "length",
+		.help = "GENEVE option data length (in 32b words)",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(UNSIGNED), item_param),
+		.args = ARGS(ARGS_ENTRY_BOUNDED(
+				struct rte_flow_item_geneve_opt, option_len,
+				0, 31)),
+	},
+	[ITEM_GENEVE_OPT_DATA] = {
+		.name = "data",
+		.help = "GENEVE option data pattern",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(HEX), item_param),
+		.args = ARGS(ARGS_ENTRY(struct rte_flow_item_geneve_opt, data),
+			     ARGS_ENTRY_ARB(0, 0),
+			     ARGS_ENTRY_ARB
+				(sizeof(struct rte_flow_item_geneve_opt),
+				ITEM_GENEVE_OPT_DATA_SIZE)),
+	},
 	/* Validate/create actions. */
 	[ACTIONS] = {
 		.name = "actions",
@@ -6482,11 +6551,14 @@ parse_hex(struct context *ctx, const struct token *token,
 	ret = snprintf(tmp, sizeof(tmp), "%u", hexlen);
 	if (ret < 0)
 		goto error;
-	push_args(ctx, arg_len);
-	ret = parse_int(ctx, token, tmp, ret, NULL, 0);
-	if (ret < 0) {
-		pop_args(ctx);
-		goto error;
+	/* Save length if requested. */
+	if (arg_len->size) {
+		push_args(ctx, arg_len);
+		ret = parse_int(ctx, token, tmp, ret, NULL, 0);
+		if (ret < 0) {
+			pop_args(ctx);
+			goto error;
+		}
 	}
 	buf = (uint8_t *)ctx->object + arg_data->offset;
 	/* Output buffer is not necessarily NUL-terminated. */
@@ -7486,6 +7558,9 @@ flow_item_default_mask(const struct rte_flow_item *item)
 	case RTE_FLOW_ITEM_TYPE_GENEVE:
 		mask = &rte_flow_item_geneve_mask;
 		break;
+	case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+		mask = &rte_flow_item_geneve_opt_mask;
+		break;
 	case RTE_FLOW_ITEM_TYPE_PPPOE_PROTO_ID:
 		mask = &rte_flow_item_pppoe_proto_id_mask;
 		break;
@@ -7598,6 +7673,8 @@ cmd_set_raw_parsed(const struct buffer *in)
 	/* process hdr from upper layer to low layer (L3/L4 -> L2). */
 	data_tail = data + ACTION_RAW_ENCAP_MAX_DATA;
 	for (i = n - 1 ; i >= 0; --i) {
+		const struct rte_flow_item_geneve_opt *opt;
+
 		item = in->args.vc.pattern + i;
 		if (item->spec == NULL)
 			item->spec = flow_item_default_mask(item);
@@ -7650,6 +7727,18 @@ cmd_set_raw_parsed(const struct buffer *in)
 		case RTE_FLOW_ITEM_TYPE_GENEVE:
 			size = sizeof(struct rte_geneve_hdr);
 			break;
+		case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+			opt = (const struct rte_flow_item_geneve_opt *)
+								item->spec;
+			size = offsetof(struct rte_flow_item_geneve_opt, data);
+			if (opt->option_len && opt->data) {
+				*total_size += opt->option_len *
+					       sizeof(uint32_t);
+				rte_memcpy(data_tail - (*total_size),
+					   opt->data,
+					   opt->option_len * sizeof(uint32_t));
+			}
+			break;
 		case RTE_FLOW_ITEM_TYPE_L2TPV3OIP:
 			size = sizeof(rte_be32_t);
 			proto = 0x73;
doc/guides/testpmd_app_ug/testpmd_funcs.rst +8 -0
index 9be450066e..37278d31d6 100644
@@ -3680,6 +3680,14 @@ This section lists supported pattern items and their attributes, if any.
   - ``vni {unsigned}``: virtual network identifier.
   - ``protocol {unsigned}``: protocol type.
 
+- ``geneve-opt``: match GENEVE header option.
+
+  - ``class {unsigned}``: GENEVE option class.
+  - ``type {unsigned}``: GENEVE option type.
+  - ``length {unsigned}``: GENEVE option length in 32-bit words.
+  - ``data {hex string}``: GENEVE option data, the legnt is defined by
+    ``length`` field
+
 - ``vxlan-gpe``: match VXLAN-GPE header.
 
   - ``vni {unsigned}``: VXLAN-GPE identifier.
Signature · 1 line
2.21.0
Acked-by on v4 2/8
#16 lore raw

Re: [dpdk-dev] [PATCH v4 2/8] app/testpmd: add GENEVE option item support

Hi,

Quoted 16 lines

-----Original Message-----
From: dev <[email protected]> On Behalf Of Shiri Kuzin
Sent: Monday, January 11, 2021 4:27 PM
Subject: [dpdk-dev] [PATCH v4 2/8] app/testpmd: add GENEVE option item support

From: Viacheslav Ovsiienko <[email protected]>

The patch adds the GENEVE option rte flow item support to command line interpreter. The flow command with GENEVE option items looks like:

flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 /
geneve-opt class is 99 length is 1 type is 0 data is 0x669988 / end actions drop / end

Quoted 8 lines

The option length should be specified in 32-bit words, this value specifies the length of the data pattern/mask arrays (should be multiplied by sizeof(uint32_t) to be expressed in bytes. If match on the length itself is not needed the mask should be set to zero, in this case length is used to specify the pattern/mask array lengths only.

---

Thanks,
Ori

#17 lore raw

PATCH v4 3/8 common/mlx5: check GENEVE TLV support in HCA attributes

In reply to Shiri Kuzin

This is preparation step to support match on GENEVE TLV option.

In this Patch we add the HCA attributes that will allow supporting GENEVE TLV option matching.

---
 drivers/common/mlx5/mlx5_devx_cmds.c |  7 +++++++
 drivers/common/mlx5/mlx5_devx_cmds.h |  4 ++++
 drivers/common/mlx5/mlx5_prm.h       | 28 +++++++++++++++++++++++++---
 3 files changed, 36 insertions(+), 3 deletions(-)
drivers/common/mlx5/mlx5_devx_cmds.c +7 -0
index 12f51a940c..fc406af062 100644
@@ -693,6 +693,10 @@ mlx5_devx_cmd_query_hca_attr(void *ctx,
 	attr->eth_virt = MLX5_GET(cmd_hca_cap, hcattr, eth_virt);
 	attr->flex_parser_protocols = MLX5_GET(cmd_hca_cap, hcattr,
 					       flex_parser_protocols);
+	attr->max_geneve_tlv_options = MLX5_GET(cmd_hca_cap, hcattr,
+			max_geneve_tlv_options);
+	attr->max_geneve_tlv_option_data_len = MLX5_GET(cmd_hca_cap, hcattr,
+			max_geneve_tlv_option_data_len);
 	attr->qos.sup = MLX5_GET(cmd_hca_cap, hcattr, qos);
 	attr->vdpa.valid = !!(MLX5_GET64(cmd_hca_cap, hcattr,
 					 general_obj_types) &
@@ -720,6 +724,9 @@ mlx5_devx_cmd_query_hca_attr(void *ctx,
 	attr->flow_hit_aso = !!(MLX5_GET64(cmd_hca_cap, hcattr,
 					   general_obj_types) &
 				MLX5_GENERAL_OBJ_TYPES_CAP_FLOW_HIT_ASO);
+	attr->geneve_tlv_opt = !!(MLX5_GET64(cmd_hca_cap, hcattr,
+					   general_obj_types) &
+				MLX5_GENERAL_OBJ_TYPES_CAP_GENEVE_TLV_OPT);
 	attr->log_max_cq = MLX5_GET(cmd_hca_cap, hcattr, log_max_cq);
 	attr->log_max_qp = MLX5_GET(cmd_hca_cap, hcattr, log_max_qp);
 	attr->log_max_cq_sz = MLX5_GET(cmd_hca_cap, hcattr, log_max_cq_sz);
drivers/common/mlx5/mlx5_devx_cmds.h +4 -0
index b335b7c82c..6b2f77837b 100644
@@ -97,6 +97,8 @@ struct mlx5_hca_attr {
 	uint32_t lro_timer_supported_periods[MLX5_LRO_NUM_SUPP_PERIODS];
 	uint16_t lro_min_mss_size;
 	uint32_t flex_parser_protocols;
+	uint32_t max_geneve_tlv_options;
+	uint32_t max_geneve_tlv_option_data_len;
 	uint32_t hairpin:1;
 	uint32_t log_max_hairpin_queues:5;
 	uint32_t log_max_hairpin_wq_data_sz:5;
@@ -116,6 +118,7 @@ struct mlx5_hca_attr {
 	uint32_t regex:1;
 	uint32_t regexp_num_of_engines;
 	uint32_t log_max_ft_sampler_num:8;
+	uint32_t geneve_tlv_opt;
 	struct mlx5_hca_qos_attr qos;
 	struct mlx5_hca_vdpa_attr vdpa;
 	int log_max_qp_sz;
@@ -479,6 +482,7 @@ struct mlx5_devx_obj *mlx5_devx_cmd_create_flex_parser(void *ctx,
 __rte_internal
 int mlx5_devx_cmd_register_read(void *ctx, uint16_t reg_id,
 				uint32_t arg, uint32_t *data, uint32_t dw_cnt);
+
 /**
  * Create virtio queue counters object DevX API.
  *
drivers/common/mlx5/mlx5_prm.h +25 -3
index 8c9b53ce10..458550279e 100644
@@ -789,7 +789,7 @@ struct mlx5_ifc_fte_match_set_misc3_bits {
 	u8 icmp_code[0x8];
 	u8 icmpv6_type[0x8];
 	u8 icmpv6_code[0x8];
-	u8 reserved_at_120[0x20];
+	u8 geneve_tlv_option_0_data[0x20];
 	u8 gtpu_teid[0x20];
 	u8 gtpu_msg_type[0x08];
 	u8 gtpu_msg_flags[0x08];
@@ -1082,6 +1082,8 @@ enum {
 			(1ULL << MLX5_GENERAL_OBJ_TYPE_FLEX_PARSE_GRAPH)
 #define MLX5_GENERAL_OBJ_TYPES_CAP_FLOW_HIT_ASO \
 			(1ULL << MLX5_GENERAL_OBJ_TYPE_FLOW_HIT_ASO)
+#define MLX5_GENERAL_OBJ_TYPES_CAP_GENEVE_TLV_OPT \
+			(1ULL << MLX5_OBJ_TYPE_GENEVE_TLV_OPT)
 
 enum {
 	MLX5_HCA_CAP_OPMOD_GET_MAX   = 0,
@@ -1380,8 +1382,10 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 	u8 reserved_at_500[0x20];
 	u8 num_of_uars_per_page[0x20];
 	u8 flex_parser_protocols[0x20];
-	u8 reserved_at_560[0x20];
-	u8 reserved_at_580[0x3c];
+	u8 max_geneve_tlv_options[0x8];
+	u8 reserved_at_568[0x3];
+	u8 max_geneve_tlv_option_data_len[0x5];
+	u8 reserved_at_570[0x4c];
 	u8 mini_cqe_resp_stride_index[0x1];
 	u8 cqe_128_always[0x1];
 	u8 cqe_compression_128[0x1];
@@ -2292,6 +2296,7 @@ struct mlx5_ifc_create_cq_in_bits {
 };
 
 enum {
+	MLX5_OBJ_TYPE_GENEVE_TLV_OPT = 0x000b,
 	MLX5_GENERAL_OBJ_TYPE_VIRTQ = 0x000d,
 	MLX5_GENERAL_OBJ_TYPE_VIRTIO_Q_COUNTERS = 0x001c,
 	MLX5_GENERAL_OBJ_TYPE_FLEX_PARSE_GRAPH = 0x0022,
@@ -2326,6 +2331,17 @@ struct mlx5_ifc_virtio_q_counters_bits {
 	u8 reserved_at_180[0x50];
 };
 
+struct mlx5_ifc_geneve_tlv_option_bits {
+	u8 modify_field_select[0x40];
+	u8 reserved_at_40[0x18];
+	u8 geneve_option_fte_index[0x8];
+	u8 option_class[0x10];
+	u8 option_type[0x8];
+	u8 reserved_at_78[0x3];
+	u8 option_data_length[0x5];
+	u8 reserved_at_80[0x180];
+};
+
 struct mlx5_ifc_create_virtio_q_counters_in_bits {
 	struct mlx5_ifc_general_obj_in_cmd_hdr_bits hdr;
 	struct mlx5_ifc_virtio_q_counters_bits virtio_q_counters;
@@ -2335,6 +2351,12 @@ struct mlx5_ifc_query_virtio_q_counters_out_bits {
 	struct mlx5_ifc_general_obj_in_cmd_hdr_bits hdr;
 	struct mlx5_ifc_virtio_q_counters_bits virtio_q_counters;
 };
+
+struct mlx5_ifc_create_geneve_tlv_option_in_bits {
+	struct mlx5_ifc_general_obj_in_cmd_hdr_bits hdr;
+	struct mlx5_ifc_geneve_tlv_option_bits geneve_tlv_opt;
+};
+
 enum {
 	MLX5_VIRTQ_STATE_INIT = 0,
 	MLX5_VIRTQ_STATE_RDY = 1,
Signature · 1 line
2.21.0
#18 lore raw

PATCH v4 4/8 common/mlx5: create GENEVE TLV option object with DevX

In reply to Shiri Kuzin

TLV object is a special firmware maintained entity used to support match on GENEVE header extension option.

The TLV object is created with DevX API and accepts the option class, type and lehgth fields.

The class type and length fields are set using MLX5_SET and the Devx object is created using mlx5 glue function.

---
 drivers/common/mlx5/mlx5_devx_cmds.c | 56 +++++++++++++++++++++++++++-
 drivers/common/mlx5/mlx5_devx_cmds.h |  5 +++
 drivers/common/mlx5/version.map      |  1 +
 3 files changed, 61 insertions(+), 1 deletion(-)
drivers/common/mlx5/mlx5_devx_cmds.c +55 -1
index fc406af062..08bd357ad1 100644
@@ -2068,7 +2068,6 @@ mlx5_devx_cmd_create_flow_hit_aso_obj(void *ctx, uint32_t pd)
  *
  * @param[in] ctx
  *   Context returned from mlx5 open_device() glue function.
- *
  * @return
  *   The DevX object created, NULL otherwise and rte_errno is set.
  */
@@ -2097,3 +2096,58 @@ mlx5_devx_cmd_alloc_pd(void *ctx)
 	ppd->id = MLX5_GET(alloc_pd_out, out, pd);
 	return ppd;
 }
+
+/**
+ * Create general object of type GENEVE TLV option using DevX API.
+ *
+ * @param[in] ctx
+ *   Context returned from mlx5 open_device() glue function.
+ * @param [in] class
+ *   TLV option variable value of class
+ * @param [in] type
+ *   TLV option variable value of type
+ * @param [in] len
+ *   TLV option variable value of len
+ *    *
+ * @return
+ *   The DevX object created, NULL otherwise and rte_errno is set.
+ */
+struct mlx5_devx_obj *
+mlx5_devx_cmd_create_geneve_tlv_option(void *ctx,
+		uint16_t class, uint8_t type, uint8_t len)
+{
+	uint32_t in[MLX5_ST_SZ_DW(create_geneve_tlv_option_in)] = {0};
+	uint32_t out[MLX5_ST_SZ_DW(general_obj_out_cmd_hdr)] = {0};
+	struct mlx5_devx_obj *geneve_tlv_opt_obj = mlx5_malloc(MLX5_MEM_ZERO,
+						   sizeof(*geneve_tlv_opt_obj),
+						   0, SOCKET_ID_ANY);
+
+	if (!geneve_tlv_opt_obj) {
+		DRV_LOG(ERR, "Failed to allocate geneve tlv option object.");
+		rte_errno = ENOMEM;
+		return NULL;
+	}
+	void *hdr = MLX5_ADDR_OF(create_geneve_tlv_option_in, in, hdr);
+	void *opt = MLX5_ADDR_OF(create_geneve_tlv_option_in, in,
+			geneve_tlv_opt);
+	MLX5_SET(general_obj_in_cmd_hdr, hdr, opcode,
+			MLX5_CMD_OP_CREATE_GENERAL_OBJECT);
+	MLX5_SET(general_obj_in_cmd_hdr, hdr, obj_type,
+			MLX5_OBJ_TYPE_GENEVE_TLV_OPT);
+	MLX5_SET(geneve_tlv_option, opt, option_class,
+			rte_be_to_cpu_16(class));
+	MLX5_SET(geneve_tlv_option, opt, option_type, type);
+	MLX5_SET(geneve_tlv_option, opt, option_data_length, len);
+	geneve_tlv_opt_obj->obj = mlx5_glue->devx_obj_create(ctx, in,
+					sizeof(in), out, sizeof(out));
+	if (!geneve_tlv_opt_obj->obj) {
+		rte_errno = errno;
+		DRV_LOG(ERR, "Failed to create Geneve tlv option "
+				"Obj using DevX.");
+		mlx5_free(geneve_tlv_opt_obj);
+		return NULL;
+	}
+	geneve_tlv_opt_obj->id = MLX5_GET(general_obj_out_cmd_hdr, out, obj_id);
+	return geneve_tlv_opt_obj;
+}
+
drivers/common/mlx5/mlx5_devx_cmds.h +5 -0
index 6b2f77837b..6774b49d2d 100644
@@ -483,6 +483,11 @@ __rte_internal
 int mlx5_devx_cmd_register_read(void *ctx, uint16_t reg_id,
 				uint32_t arg, uint32_t *data, uint32_t dw_cnt);
 
+__rte_internal
+struct mlx5_devx_obj *
+mlx5_devx_cmd_create_geneve_tlv_option(void *ctx,
+		uint16_t class, uint8_t type, uint8_t len);
+
 /**
  * Create virtio queue counters object DevX API.
  *
drivers/common/mlx5/version.map +1 -0
index fb3952bbc6..7de8006b89 100644
@@ -23,6 +23,7 @@ INTERNAL {
 	mlx5_devx_cmd_create_virtio_q_counters;
 	mlx5_devx_cmd_create_virtq;
         mlx5_devx_cmd_create_flow_hit_aso_obj;
+	mlx5_devx_cmd_create_geneve_tlv_option;
 	mlx5_devx_cmd_destroy;
 	mlx5_devx_cmd_flow_counter_alloc;
 	mlx5_devx_cmd_flow_counter_query;
Signature · 1 line
2.21.0
#19 lore raw

PATCH v4 5/8 net/mlx5: create GENEVE TLV option management

In reply to Shiri Kuzin

Currently firmware supports the only TLV object per device to match on the GENEVE header option.

This patch adds the simple TLV object management to the mlx5 PMD.

---
 drivers/net/mlx5/mlx5.c         |   2 +
 drivers/net/mlx5/mlx5.h         |  13 ++++
 drivers/net/mlx5/mlx5_flow.h    |   4 ++
 drivers/net/mlx5/mlx5_flow_dv.c | 108 ++++++++++++++++++++++++++++++++
 4 files changed, 127 insertions(+)
drivers/net/mlx5/mlx5.c +2 -0
index 023ef50a77..68d6352d48 100644
@@ -1014,6 +1014,7 @@ mlx5_alloc_shared_dev_ctx(const struct mlx5_dev_spawn_data *spawn,
 	rte_rwlock_write_unlock(&mlx5_shared_data->mem_event_rwlock);
 	/* Add context to the global device list. */
 	LIST_INSERT_HEAD(&mlx5_dev_ctx_list, sh, next);
+	rte_spinlock_init(&sh->geneve_tlv_opt_sl);
 exit:
 	pthread_mutex_unlock(&mlx5_dev_ctx_list_mutex);
 	return sh;
@@ -1109,6 +1110,7 @@ mlx5_free_shared_dev_ctx(struct mlx5_dev_ctx_shared *sh)
 		mlx5_glue->devx_free_uar(sh->devx_rx_uar);
 	if (sh->ctx)
 		claim_zero(mlx5_glue->close_device(sh->ctx));
+	MLX5_ASSERT(sh->geneve_tlv_option_resource == NULL);
 	pthread_mutex_destroy(&sh->txpp.mutex);
 	mlx5_free(sh);
 	return;
drivers/net/mlx5/mlx5.h +13 -0
index 41034f5d19..23272950a4 100644
@@ -540,6 +540,16 @@ struct mlx5_aso_age_mng {
 	struct mlx5_aso_sq aso_sq; /* ASO queue objects. */
 };
 
+/* Management structure for geneve tlv option */
+struct mlx5_geneve_tlv_option_resource {
+	struct mlx5_devx_obj *obj; /* Pointer to the geneve tlv opt object. */
+	rte_be16_t option_class; /* geneve tlv opt class.*/
+	uint8_t option_type; /* geneve tlv opt type.*/
+	uint8_t length; /* geneve tlv opt length. */
+	uint32_t refcnt; /* geneve tlv object reference counter */
+};
+
+
 #define MLX5_AGE_EVENT_NEW		1
 #define MLX5_AGE_TRIGGER		2
 #define MLX5_AGE_SET(age_info, BIT) \
@@ -752,6 +762,9 @@ struct mlx5_dev_ctx_shared {
 	void *devx_rx_uar; /* DevX UAR for Rx. */
 	struct mlx5_aso_age_mng *aso_age_mng;
 	/* Management data for aging mechanism using ASO Flow Hit. */
+	struct mlx5_geneve_tlv_option_resource *geneve_tlv_option_resource;
+	/* Management structure for geneve tlv option */
+	rte_spinlock_t geneve_tlv_opt_sl; /* Lock for geneve tlv resource */
 	struct mlx5_dev_shared_port port[]; /* per device port data array. */
 };
 
drivers/net/mlx5/mlx5_flow.h +4 -0
index ee85c9d8a5..0c8861964c 100644
@@ -1046,6 +1046,7 @@ struct rte_flow {
 	uint32_t counter; /**< Holds flow counter. */
 	uint32_t tunnel_id;  /**< Tunnel id */
 	uint32_t age; /**< Holds ASO age bit index. */
+	uint32_t geneve_tlv_option; /**< Holds Geneve TLV option id. > */
 } __rte_packed;
 
 /*
@@ -1503,4 +1504,7 @@ void flow_dv_dest_array_remove_cb(struct mlx5_cache_list *list,
 				  struct mlx5_cache_entry *entry);
 struct mlx5_aso_age_action *flow_aso_age_get_by_idx(struct rte_eth_dev *dev,
 						    uint32_t age_idx);
+int flow_dev_geneve_tlv_option_resource_register(struct rte_eth_dev *dev,
+					     const struct rte_flow_item *item,
+					     struct rte_flow_error *error);
 #endif /* RTE_PMD_MLX5_FLOW_H_ */
drivers/net/mlx5/mlx5_flow_dv.c +108 -0

126 lines not on this page · raw message

Signature · 1 line
2.21.0
#20 lore raw

PATCH v4 6/8 net/mlx5: add GENEVE TLV option flow validation

In reply to Shiri Kuzin

This patch adds validation routine for the GENEVE header TLV option.

The GENEVE TLV option match must include all fields with full masks due to NIC does not support masking on option class, type and length.

The option data length must be non zero and provided data pattern should be zero neither due to hardware limitations.

---
 drivers/net/mlx5/mlx5_flow.c    | 120 ++++++++++++++++++++++++++++++++
 drivers/net/mlx5/mlx5_flow.h    |   7 ++
 drivers/net/mlx5/mlx5_flow_dv.c |  10 ++-
 3 files changed, 136 insertions(+), 1 deletion(-)
drivers/net/mlx5/mlx5_flow.c +120 -0
index 2a4073c126..892ed2f2d9 100644
@@ -2627,6 +2627,126 @@ mlx5_flow_validate_item_geneve(const struct rte_flow_item *item,
 	return 0;
 }
 
+/**
+ * Validate Geneve TLV option item.
+ *
+ * @param[in] item
+ *   Item specification.
+ * @param[in] last_item
+ *   Previous validated item in the pattern items.
+ * @param[in] dev
+ *   Pointer to the rte_eth_dev structure.
+ * @param[out] error
+ *   Pointer to error structure.
+ *
+ * @return
+ *   0 on success, a negative errno value otherwise and rte_errno is set.
+ */
+int
+mlx5_flow_validate_item_geneve_opt(const struct rte_flow_item *item,
+				   uint64_t last_item,
+				   struct rte_eth_dev *dev,
+				   struct rte_flow_error *error)
+{
+	struct mlx5_priv *priv = dev->data->dev_private;
+	struct mlx5_dev_ctx_shared *sh = priv->sh;
+	struct mlx5_geneve_tlv_option_resource *geneve_opt_resource;
+	struct mlx5_hca_attr *hca_attr = &priv->config.hca_attr;
+	uint8_t data_max_supported =
+			hca_attr->max_geneve_tlv_option_data_len * 4;
+	struct mlx5_dev_config *config = &priv->config;
+	const struct rte_flow_item_geneve_opt *spec = item->spec;
+	const struct rte_flow_item_geneve_opt *mask = item->mask;
+	unsigned int i;
+	unsigned int data_len;
+	const struct rte_flow_item_geneve_opt full_mask = {
+		.option_class = RTE_BE16(0xffff),
+		.option_type = 0xff,
+		.option_len = 0x1f,
+	};
+
+	if (!mask)
+		mask = &rte_flow_item_geneve_opt_mask;
+	if (!spec)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt class/type/length must be specified");
+	if ((uint32_t)(spec->option_len) > MLX5_GENEVE_OPTLEN_MASK)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt length exceeeds the limit (31)");
+	/* Check if class type and length masks are full. */
+	if (full_mask.option_class != mask->option_class ||
+	    full_mask.option_type != mask->option_type ||
+	    full_mask.option_len != (mask->option_len & full_mask.option_len))
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt class/type/length masks must be full");
+	/* Check if length is supported */
+	if ((uint32_t)(spec->option_len) >
+			config->hca_attr.max_geneve_tlv_option_data_len)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt length not supported");
+	if (config->hca_attr.max_geneve_tlv_options > 1)
+		DRV_LOG(DEBUG,
+			"max_geneve_tlv_options supports more than 1 option");
+	/* Check GENEVE item preceding. */
+	if (!(last_item & MLX5_FLOW_LAYER_GENEVE))
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve opt item must be preceded with Geneve item");
+	/* Check if length is 0 or data is 0. */
+	if (spec->data == NULL || spec->option_len == 0)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt with zero data/length not supported");
+	/* Check not all data & mask are 0. */
+	data_len = spec->option_len * 4;
+	if (mask->data == NULL) {
+		for (i = 0; i < data_len; i++)
+			if (spec->data[i])
+				break;
+		if (i == data_len)
+			return rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_ITEM, item,
+				"Can't match on Geneve option data 0");
+	} else {
+		for (i = 0; i < data_len; i++)
+			if (spec->data[i] & mask->data[i])
+				break;
+		if (i == data_len)
+			return rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_ITEM, item,
+				"Can't match on Geneve option data and mask 0");
+		/* Check data mask supported. */
+		for (i = data_max_supported; i < data_len ; i++)
+			if (mask->data[i])
+				return rte_flow_error_set(error, ENOTSUP,
+					RTE_FLOW_ERROR_TYPE_ITEM, item,
+					"Data mask is of unsupported size");
+	}
+	/* Check GENEVE option is supported in NIC. */
+	if (!config->hca_attr.geneve_tlv_opt)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt not supported");
+	/* Check if we already have geneve option with different type/class. */
+	rte_spinlock_lock(&sh->geneve_tlv_opt_sl);
+	geneve_opt_resource = sh->geneve_tlv_option_resource;
+	if (geneve_opt_resource != NULL)
+		if (geneve_opt_resource->option_class != spec->option_class ||
+		    geneve_opt_resource->option_type != spec->option_type ||
+		    geneve_opt_resource->length != spec->option_len) {
+			rte_spinlock_unlock(&sh->geneve_tlv_opt_sl);
+			return rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_ITEM, item,
+				"Only one Geneve TLV option supported");
+		}
+	rte_spinlock_unlock(&sh->geneve_tlv_opt_sl);
+	return 0;
+}
+
 /**
  * Validate MPLS item.
  *
drivers/net/mlx5/mlx5_flow.h +7 -0
index 0c8861964c..28db5faee1 100644
@@ -138,6 +138,9 @@ enum mlx5_feature_name {
 #define MLX5_FLOW_LAYER_OUTER_L3_IPV6_FRAG_EXT (1u << 30)
 #define MLX5_FLOW_LAYER_INNER_L3_IPV6_FRAG_EXT (1u << 31)
 
+/* Pattern tunnel Layer bits (continued). */
+#define MLX5_FLOW_LAYER_GENEVE_OPT (UINT64_C(1) << 32)
+
 /* Outer Masks. */
 #define MLX5_FLOW_LAYER_OUTER_L3 \
 	(MLX5_FLOW_LAYER_OUTER_L3_IPV4 | MLX5_FLOW_LAYER_OUTER_L3_IPV6)
@@ -1396,6 +1399,10 @@ int mlx5_flow_validate_item_geneve(const struct rte_flow_item *item,
 				   uint64_t item_flags,
 				   struct rte_eth_dev *dev,
 				   struct rte_flow_error *error);
+int mlx5_flow_validate_item_geneve_opt(const struct rte_flow_item *item,
+				   uint64_t last_item,
+				   struct rte_eth_dev *dev,
+				   struct rte_flow_error *error);
 int mlx5_flow_validate_item_ecpri(const struct rte_flow_item *item,
 				  uint64_t item_flags,
 				  uint64_t last_item,
drivers/net/mlx5/mlx5_flow_dv.c +9 -1
index 0a657cea41..c24123fbf3 100644
@@ -5520,6 +5520,15 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr,
 				return ret;
 			last_item = MLX5_FLOW_LAYER_GENEVE;
 			break;
+		case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+			ret = mlx5_flow_validate_item_geneve_opt(items,
+								 last_item,
+								 dev,
+								 error);
+			if (ret < 0)
+				return ret;
+			last_item = MLX5_FLOW_LAYER_GENEVE_OPT;
+			break;
 		case RTE_FLOW_ITEM_TYPE_MPLS:
 			ret = mlx5_flow_validate_item_mpls(dev, items,
 							   item_flags,
@@ -5528,7 +5537,6 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr,
 				return ret;
 			last_item = MLX5_FLOW_LAYER_MPLS;
 			break;
-
 		case RTE_FLOW_ITEM_TYPE_MARK:
 			ret = flow_dv_validate_item_mark(dev, items, attr,
 							 error);
Signature · 1 line
2.21.0
#21 lore raw

PATCH v4 7/8 net/mlx5: add GENEVE TLV option flow translation

In reply to Shiri Kuzin

The GENEVE TLV option matching flows must be created using a translation function.

This function checks whether we already created a Devx object for the matching and either creates the objects or updates the reference counter.

---
 drivers/net/mlx5/mlx5_flow.h    |  2 +-
 drivers/net/mlx5/mlx5_flow_dv.c | 80 +++++++++++++++++++++++++++++++++
 2 files changed, 81 insertions(+), 1 deletion(-)
drivers/net/mlx5/mlx5_flow.h +1 -1
index 28db5faee1..d6a6839476 100644
@@ -329,7 +329,7 @@ enum mlx5_feature_name {
 #define MLX5_GENEVE_VER_VAL(a) \
 		(((a) >> (MLX5_GENEVE_VER_SHIFT)) & (MLX5_GENEVE_VER_MASK))
 #define MLX5_GENEVE_OPTLEN_MASK 0x3F
-#define MLX5_GENEVE_OPTLEN_SHIFT 7
+#define MLX5_GENEVE_OPTLEN_SHIFT 8
 #define MLX5_GENEVE_OPTLEN_VAL(a) \
 	    (((a) >> (MLX5_GENEVE_OPTLEN_SHIFT)) & (MLX5_GENEVE_OPTLEN_MASK))
 #define MLX5_GENEVE_OAMF_MASK 0x1
drivers/net/mlx5/mlx5_flow_dv.c +80 -0

92 lines not on this page · raw message

Signature · 1 line
2.21.0
#22 lore raw

PATCH v4 8/8 doc: update GENEVE TLV option support

In reply to Shiri Kuzin

GENEVE TLV option support added to mlx5 PMD.

The limitations and support were updated in
documentation.

---
 doc/guides/nics/mlx5.rst               | 23 ++++++++++++++++++++++-
 doc/guides/rel_notes/release_21_02.rst |  8 ++++++++
 2 files changed, 30 insertions(+), 1 deletion(-)
doc/guides/nics/mlx5.rst +22 -1
index 3bda0f8417..5f2b42b979 100644
@@ -98,6 +98,7 @@ Features
 - Hardware LRO.
 - Hairpin.
 - Multiple-thread flow insertion.
+- Matching on Geneve TLV option header with raw encap/decap action.
 
 Limitations
 -----------
@@ -175,7 +176,18 @@ Limitations
      - OAM
      - protocol type
      - options length
-       Currently, the only supported options length value is 0.
+
+- Match on Geneve TLv option is supported on the following fields:
+     - Class
+     - Type
+     - Length
+     - Data
+
+  Only one Class/Type/Length Geneve TLV option is supported per shared device.
+  Class/Type/Length fields must be specified as well as masks.
+  Class/Type/Length specified masks must be full.
+  Matching Geneve TLV option without specifying data is not supported.
+  Matching Geneve TLV option with data & mask == 0 is not supported.
 
 - VF: flow rules created on VF devices can only match traffic targeted at the
   configured MAC addresses (see ``rte_eth_dev_mac_addr_add()``).
@@ -1022,6 +1034,10 @@ Below are some firmware configurations listed.
    or
    FLEX_PARSER_PROFILE_ENABLE=1
 
+- enable Geneve TLV option flow matching::
+
+   FLEX_PARSER_PROFILE_ENABLE=0
+
 - enable GTP flow matching::
 
    FLEX_PARSER_PROFILE_ENABLE=3
@@ -1501,6 +1517,11 @@ Supported hardware offloads
    |                       | |  rdma-core 32 | | rdma-core 32  |
    |                       | |  ConnectX-6 Dx| | ConnectX-6 Dx |
    +-----------------------+-----------------+-----------------+
+   | Encapsulation         | | DPDK 21.02    | | DPDK 21.02    |
+   | GENEVE TLV option     | | OFED 5.2      | | OFED 5.2      |
+   |                       | | rdma-core 34  | | rdma-core 34  |
+   |                       | | ConnectX-6 Dx | | ConnectX-6 Dx |
+   +-----------------------+-----------------+-----------------+
 
 Notes for metadata
 ------------------
doc/guides/rel_notes/release_21_02.rst +8 -0
index 638f98168b..2fb5bf9c38 100644
@@ -55,6 +55,14 @@ New Features
      Also, make sure to start the actual text at the margin.
      =======================================================
 
+* **Updated Mellanox mlx5 driver.**
+
+  Updated the Mellanox mlx5 driver with new features and improvements, including:
+
+  * **Added GENEVE TLV option in rte_flow.**
+
+  Added support for matching GENEVE TLV option and raw encap/decap of GENEVE
+  TLV option.
 
 Removed Items
 -------------
Signature · 1 line
2.21.0
#23 lore raw

v5 cover ethdev: introduce GENEVE header TLV option item

In reply to Shiri Kuzin

The Geneve tunneling protocol is designed to allow the user to specify some data context on the packet.
The GENEVE TLV (Type-Length-Variable) Option is the mean intended to present the user data.

In order to support GENEVE TLV Option the new rte_flow item "rte_flow_item_geneve_opt" is introduced.
The new item contains the values and masks for the following fields:
-option class
-option type
-length
-data

The usage example:
"flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 / geneve-opt class is 5 length is 1 type is 0 data is 0x66998800 / end actions count / drop / end"

New item will be added to testpmd to support raw encap/decap action.

v5:
-added Devx function to def file.

v4:
- fixed setting option length.
- fixed testpmd calculation of GENEVE option size.
- updated documentation.

v3:
- updated documentation.

v2:
- removed pedantic.

Shiri Kuzin (7):
lib/librte_ethdev: introduce GENEVE header TLV option item common/mlx5: check GENEVE TLV support in HCA attributes common/mlx5: create GENEVE TLV option object with DevX net/mlx5: create GENEVE TLV option management net/mlx5: add GENEVE TLV option flow validation net/mlx5: add GENEVE TLV option flow translation
doc: update GENEVE TLV option support

Viacheslav Ovsiienko (1):
app/testpmd: add GENEVE option item support

 app/test-pmd/cmdline_flow.c                   | 101 ++++++++-
 doc/guides/nics/mlx5.rst                      |  23 +-
 doc/guides/rel_notes/release_21_02.rst        |   8 +
 doc/guides/testpmd_app_ug/testpmd_funcs.rst   |   8 +
 drivers/common/mlx5/mlx5_devx_cmds.c          |  63 +++++-
 drivers/common/mlx5/mlx5_devx_cmds.h          |   9 +
 drivers/common/mlx5/mlx5_prm.h                |  28 ++-
 .../common/mlx5/rte_common_mlx5_exports.def   |   1 +
 drivers/common/mlx5/version.map               |   1 +
 drivers/net/mlx5/mlx5.c                       |   2 +
 drivers/net/mlx5/mlx5.h                       |  13 ++
 drivers/net/mlx5/mlx5_flow.c                  | 120 +++++++++++
 drivers/net/mlx5/mlx5_flow.h                  |  13 +-
 drivers/net/mlx5/mlx5_flow_dv.c               | 198 +++++++++++++++++-
 lib/librte_ethdev/rte_flow.c                  |   1 +
 lib/librte_ethdev/rte_flow.h                  |  27 +++
 16 files changed, 603 insertions(+), 13 deletions(-)
Signature · 1 line
2.21.0
#24 lore raw

PATCH v5 1/8 lib/librte_ethdev: introduce GENEVE header TLV option item

The Geneve tunneling protocol is designed to allow the user to specify some data context on the packet. The GENEVE TLV (Type-Length-Variable) Option is the mean intended to present the user data.

In order to support GENEVE TLV Option the new rte_flow item "rte_flow_item_geneve_opt" is added.
The new item contains the values and masks for the following fields:
-option class
-option type
-length
-data

New item will be added to testpmd to support match and raw encap/decap actions.

---
 lib/librte_ethdev/rte_flow.c |  1 +
 lib/librte_ethdev/rte_flow.h | 27 +++++++++++++++++++++++++++
 2 files changed, 28 insertions(+)
lib/librte_ethdev/rte_flow.c +1 -0
index a06f64c271..2af7d965e1 100644
@@ -97,6 +97,7 @@ static const struct rte_flow_desc_data rte_flow_desc_item[] = {
 	MK_FLOW_ITEM(L2TPV3OIP, sizeof(struct rte_flow_item_l2tpv3oip)),
 	MK_FLOW_ITEM(PFCP, sizeof(struct rte_flow_item_pfcp)),
 	MK_FLOW_ITEM(ECPRI, sizeof(struct rte_flow_item_ecpri)),
+	MK_FLOW_ITEM(GENEVE_OPT, sizeof(struct rte_flow_item_geneve_opt)),
 };
 
 /** Generate flow_action[] entry. */
lib/librte_ethdev/rte_flow.h +27 -0
index 0977a78270..11a6494b8e 100644
@@ -543,6 +543,14 @@ enum rte_flow_item_type {
 	 * See struct rte_flow_item_ipv6_frag_ext.
 	 */
 	RTE_FLOW_ITEM_TYPE_IPV6_FRAG_EXT,
+
+	/**
+	 * Matches Geneve Variable Length Option
+	 *
+	 * See struct rte_flow_item_geneve_opt
+	 */
+	RTE_FLOW_ITEM_TYPE_GENEVE_OPT,
+
 };
 
 /**
@@ -1627,6 +1635,25 @@ static const struct rte_flow_item_ecpri rte_flow_item_ecpri_mask = {
 };
 #endif
 
+/**
+ * RTE_FLOW_ITEM_TYPE_GENEVE_OPT
+ *
+ * Matches a GENEVE Variable Length Option
+ */
+struct rte_flow_item_geneve_opt {
+	rte_be16_t option_class;
+	uint8_t option_type;
+	uint8_t option_len;
+	uint32_t *data;
+};
+
+/** Default mask for RTE_FLOW_ITEM_TYPE_GENEVE_OPT. */
+#ifndef __cplusplus
+static const struct rte_flow_item_geneve_opt
+rte_flow_item_geneve_opt_mask = {
+	.option_type = 0xff,
+};
+#endif
 /**
  * Matching pattern item definition.
  *
Signature · 1 line
2.21.0
#25 lore raw

PATCH v5 2/8 app/testpmd: add GENEVE option item support

In reply to Shiri Kuzin

From: Viacheslav Ovsiienko <[email protected]>

The patch adds the GENEVE option rte flow item support to command line interpreter. The flow command with GENEVE option items looks like:

flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 /
geneve-opt class is 99 length is 1 type is 0 data is 0x669988 / end actions drop / end

The option length should be specified in 32-bit words, this value specifies the length of the data pattern/mask arrays (should be multiplied by sizeof(uint32_t) to be expressed in bytes. If match on the length itself is not needed the mask should be set to zero, in this case length is used to specify the pattern/mask array lengths only.

---
 app/test-pmd/cmdline_flow.c                 | 101 ++++++++++++++++++--
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |   8 ++
 2 files changed, 103 insertions(+), 6 deletions(-)
app/test-pmd/cmdline_flow.c +95 -6
index 585cab98b4..98afde82cc 100644
@@ -283,6 +283,11 @@ enum index {
 	ITEM_ECPRI_MSG_IQ_DATA_PCID,
 	ITEM_ECPRI_MSG_RTC_CTRL_RTCID,
 	ITEM_ECPRI_MSG_DLY_MSR_MSRID,
+	ITEM_GENEVE_OPT,
+	ITEM_GENEVE_OPT_CLASS,
+	ITEM_GENEVE_OPT_TYPE,
+	ITEM_GENEVE_OPT_LENGTH,
+	ITEM_GENEVE_OPT_DATA,
 
 	/* Validate/create actions. */
 	ACTIONS,
@@ -413,6 +418,9 @@ enum index {
 /** Maximum size for pattern in struct rte_flow_item_raw. */
 #define ITEM_RAW_PATTERN_SIZE 40
 
+/** Maximum size for GENEVE option data pattern in bytes. */
+#define ITEM_GENEVE_OPT_DATA_SIZE 124
+
 /** Storage size for struct rte_flow_item_raw including pattern. */
 #define ITEM_RAW_SIZE \
 	(sizeof(struct rte_flow_item_raw) + ITEM_RAW_PATTERN_SIZE)
@@ -428,7 +436,7 @@ struct action_rss_data {
 };
 
 /** Maximum data size in struct rte_flow_action_raw_encap. */
-#define ACTION_RAW_ENCAP_MAX_DATA 128
+#define ACTION_RAW_ENCAP_MAX_DATA 512
 #define RAW_ENCAP_CONFS_MAX_NUM 8
 
 /** Storage for struct rte_flow_action_raw_encap. */
@@ -658,6 +666,16 @@ struct token {
 		.mask = (const void *)&(const s){ .f = (1 << (b)) - 1 }, \
 	})
 
+/** Static initializer for ARGS() to target a field with limits. */
+#define ARGS_ENTRY_BOUNDED(s, f, i, a) \
+	(&(const struct arg){ \
+		.bounded = 1, \
+		.min = (i), \
+		.max = (a), \
+		.offset = offsetof(s, f), \
+		.size = sizeof(((s *)0)->f), \
+	})
+
 /** Static initializer for ARGS() to target an arbitrary bit-mask. */
 #define ARGS_ENTRY_MASK(s, f, m) \
 	(&(const struct arg){ \
@@ -903,6 +921,7 @@ static const enum index next_item[] = {
 	ITEM_AH,
 	ITEM_PFCP,
 	ITEM_ECPRI,
+	ITEM_GENEVE_OPT,
 	END_SET,
 	ZERO,
 };
@@ -1244,6 +1263,15 @@ static const enum index item_ecpri_common_type[] = {
 	ZERO,
 };
 
+static const enum index item_geneve_opt[] = {
+	ITEM_GENEVE_OPT_CLASS,
+	ITEM_GENEVE_OPT_TYPE,
+	ITEM_GENEVE_OPT_LENGTH,
+	ITEM_GENEVE_OPT_DATA,
+	ITEM_NEXT,
+	ZERO,
+};
+
 static const enum index next_action[] = {
 	ACTION_END,
 	ACTION_VOID,
@@ -3230,6 +3258,47 @@ static const struct token token_list[] = {
 		.args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_ecpri,
 				hdr.type5.msr_id)),
 	},
+	[ITEM_GENEVE_OPT] = {
+		.name = "geneve-opt",
+		.help = "GENEVE header option",
+		.priv = PRIV_ITEM(GENEVE_OPT,
+				  sizeof(struct rte_flow_item_geneve_opt) +
+				  ITEM_GENEVE_OPT_DATA_SIZE),
+		.next = NEXT(item_geneve_opt),
+		.call = parse_vc,
+	},
+	[ITEM_GENEVE_OPT_CLASS]	= {
+		.name = "class",
+		.help = "GENEVE option class",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(UNSIGNED), item_param),
+		.args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_geneve_opt,
+					     option_class)),
+	},
+	[ITEM_GENEVE_OPT_TYPE] = {
+		.name = "type",
+		.help = "GENEVE option type",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(UNSIGNED), item_param),
+		.args = ARGS(ARGS_ENTRY(struct rte_flow_item_geneve_opt,
+					option_type)),
+	},
+	[ITEM_GENEVE_OPT_LENGTH] = {
+		.name = "length",
+		.help = "GENEVE option data length (in 32b words)",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(UNSIGNED), item_param),
+		.args = ARGS(ARGS_ENTRY_BOUNDED(
+				struct rte_flow_item_geneve_opt, option_len,
+				0, 31)),
+	},
+	[ITEM_GENEVE_OPT_DATA] = {
+		.name = "data",
+		.help = "GENEVE option data pattern",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(HEX), item_param),
+		.args = ARGS(ARGS_ENTRY(struct rte_flow_item_geneve_opt, data),
+			     ARGS_ENTRY_ARB(0, 0),
+			     ARGS_ENTRY_ARB
+				(sizeof(struct rte_flow_item_geneve_opt),
+				ITEM_GENEVE_OPT_DATA_SIZE)),
+	},
 	/* Validate/create actions. */
 	[ACTIONS] = {
 		.name = "actions",
@@ -6482,11 +6551,14 @@ parse_hex(struct context *ctx, const struct token *token,
 	ret = snprintf(tmp, sizeof(tmp), "%u", hexlen);
 	if (ret < 0)
 		goto error;
-	push_args(ctx, arg_len);
-	ret = parse_int(ctx, token, tmp, ret, NULL, 0);
-	if (ret < 0) {
-		pop_args(ctx);
-		goto error;
+	/* Save length if requested. */
+	if (arg_len->size) {
+		push_args(ctx, arg_len);
+		ret = parse_int(ctx, token, tmp, ret, NULL, 0);
+		if (ret < 0) {
+			pop_args(ctx);
+			goto error;
+		}
 	}
 	buf = (uint8_t *)ctx->object + arg_data->offset;
 	/* Output buffer is not necessarily NUL-terminated. */
@@ -7486,6 +7558,9 @@ flow_item_default_mask(const struct rte_flow_item *item)
 	case RTE_FLOW_ITEM_TYPE_GENEVE:
 		mask = &rte_flow_item_geneve_mask;
 		break;
+	case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+		mask = &rte_flow_item_geneve_opt_mask;
+		break;
 	case RTE_FLOW_ITEM_TYPE_PPPOE_PROTO_ID:
 		mask = &rte_flow_item_pppoe_proto_id_mask;
 		break;
@@ -7598,6 +7673,8 @@ cmd_set_raw_parsed(const struct buffer *in)
 	/* process hdr from upper layer to low layer (L3/L4 -> L2). */
 	data_tail = data + ACTION_RAW_ENCAP_MAX_DATA;
 	for (i = n - 1 ; i >= 0; --i) {
+		const struct rte_flow_item_geneve_opt *opt;
+
 		item = in->args.vc.pattern + i;
 		if (item->spec == NULL)
 			item->spec = flow_item_default_mask(item);
@@ -7650,6 +7727,18 @@ cmd_set_raw_parsed(const struct buffer *in)
 		case RTE_FLOW_ITEM_TYPE_GENEVE:
 			size = sizeof(struct rte_geneve_hdr);
 			break;
+		case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+			opt = (const struct rte_flow_item_geneve_opt *)
+								item->spec;
+			size = offsetof(struct rte_flow_item_geneve_opt, data);
+			if (opt->option_len && opt->data) {
+				*total_size += opt->option_len *
+					       sizeof(uint32_t);
+				rte_memcpy(data_tail - (*total_size),
+					   opt->data,
+					   opt->option_len * sizeof(uint32_t));
+			}
+			break;
 		case RTE_FLOW_ITEM_TYPE_L2TPV3OIP:
 			size = sizeof(rte_be32_t);
 			proto = 0x73;
doc/guides/testpmd_app_ug/testpmd_funcs.rst +8 -0
index 9be450066e..37278d31d6 100644
@@ -3680,6 +3680,14 @@ This section lists supported pattern items and their attributes, if any.
   - ``vni {unsigned}``: virtual network identifier.
   - ``protocol {unsigned}``: protocol type.
 
+- ``geneve-opt``: match GENEVE header option.
+
+  - ``class {unsigned}``: GENEVE option class.
+  - ``type {unsigned}``: GENEVE option type.
+  - ``length {unsigned}``: GENEVE option length in 32-bit words.
+  - ``data {hex string}``: GENEVE option data, the legnt is defined by
+    ``length`` field
+
 - ``vxlan-gpe``: match VXLAN-GPE header.
 
   - ``vni {unsigned}``: VXLAN-GPE identifier.
Signature · 1 line
2.21.0
#26 lore raw

PATCH v5 3/8 common/mlx5: check GENEVE TLV support in HCA attributes

In reply to Shiri Kuzin

This is preparation step to support match on GENEVE TLV option.

In this Patch we add the HCA attributes that will allow supporting GENEVE TLV option matching.

---
 drivers/common/mlx5/mlx5_devx_cmds.c |  7 +++++++
 drivers/common/mlx5/mlx5_devx_cmds.h |  4 ++++
 drivers/common/mlx5/mlx5_prm.h       | 28 +++++++++++++++++++++++++---
 3 files changed, 36 insertions(+), 3 deletions(-)
drivers/common/mlx5/mlx5_devx_cmds.c +7 -0
index 3bf5279ff1..ad54432eef 100644
@@ -693,6 +693,10 @@ mlx5_devx_cmd_query_hca_attr(void *ctx,
 	attr->eth_virt = MLX5_GET(cmd_hca_cap, hcattr, eth_virt);
 	attr->flex_parser_protocols = MLX5_GET(cmd_hca_cap, hcattr,
 					       flex_parser_protocols);
+	attr->max_geneve_tlv_options = MLX5_GET(cmd_hca_cap, hcattr,
+			max_geneve_tlv_options);
+	attr->max_geneve_tlv_option_data_len = MLX5_GET(cmd_hca_cap, hcattr,
+			max_geneve_tlv_option_data_len);
 	attr->qos.sup = MLX5_GET(cmd_hca_cap, hcattr, qos);
 	attr->vdpa.valid = !!(MLX5_GET64(cmd_hca_cap, hcattr,
 					 general_obj_types) &
@@ -720,6 +724,9 @@ mlx5_devx_cmd_query_hca_attr(void *ctx,
 	attr->flow_hit_aso = !!(MLX5_GET64(cmd_hca_cap, hcattr,
 					   general_obj_types) &
 				MLX5_GENERAL_OBJ_TYPES_CAP_FLOW_HIT_ASO);
+	attr->geneve_tlv_opt = !!(MLX5_GET64(cmd_hca_cap, hcattr,
+					   general_obj_types) &
+				MLX5_GENERAL_OBJ_TYPES_CAP_GENEVE_TLV_OPT);
 	attr->log_max_cq = MLX5_GET(cmd_hca_cap, hcattr, log_max_cq);
 	attr->log_max_qp = MLX5_GET(cmd_hca_cap, hcattr, log_max_qp);
 	attr->log_max_cq_sz = MLX5_GET(cmd_hca_cap, hcattr, log_max_cq_sz);
drivers/common/mlx5/mlx5_devx_cmds.h +4 -0
index 94e9bbb906..2286569f37 100644
@@ -97,6 +97,8 @@ struct mlx5_hca_attr {
 	uint32_t lro_timer_supported_periods[MLX5_LRO_NUM_SUPP_PERIODS];
 	uint16_t lro_min_mss_size;
 	uint32_t flex_parser_protocols;
+	uint32_t max_geneve_tlv_options;
+	uint32_t max_geneve_tlv_option_data_len;
 	uint32_t hairpin:1;
 	uint32_t log_max_hairpin_queues:5;
 	uint32_t log_max_hairpin_wq_data_sz:5;
@@ -116,6 +118,7 @@ struct mlx5_hca_attr {
 	uint32_t regex:1;
 	uint32_t regexp_num_of_engines;
 	uint32_t log_max_ft_sampler_num:8;
+	uint32_t geneve_tlv_opt;
 	struct mlx5_hca_qos_attr qos;
 	struct mlx5_hca_vdpa_attr vdpa;
 	int log_max_qp_sz;
@@ -482,6 +485,7 @@ struct mlx5_devx_obj *mlx5_devx_cmd_create_flex_parser(void *ctx,
 __rte_internal
 int mlx5_devx_cmd_register_read(void *ctx, uint16_t reg_id,
 				uint32_t arg, uint32_t *data, uint32_t dw_cnt);
+
 /**
  * Create virtio queue counters object DevX API.
  *
drivers/common/mlx5/mlx5_prm.h +25 -3
index 7d5cf961b1..e6cf734872 100644
@@ -789,7 +789,7 @@ struct mlx5_ifc_fte_match_set_misc3_bits {
 	u8 icmp_code[0x8];
 	u8 icmpv6_type[0x8];
 	u8 icmpv6_code[0x8];
-	u8 reserved_at_120[0x20];
+	u8 geneve_tlv_option_0_data[0x20];
 	u8 gtpu_teid[0x20];
 	u8 gtpu_msg_type[0x08];
 	u8 gtpu_msg_flags[0x08];
@@ -1082,6 +1082,8 @@ enum {
 			(1ULL << MLX5_GENERAL_OBJ_TYPE_FLEX_PARSE_GRAPH)
 #define MLX5_GENERAL_OBJ_TYPES_CAP_FLOW_HIT_ASO \
 			(1ULL << MLX5_GENERAL_OBJ_TYPE_FLOW_HIT_ASO)
+#define MLX5_GENERAL_OBJ_TYPES_CAP_GENEVE_TLV_OPT \
+			(1ULL << MLX5_OBJ_TYPE_GENEVE_TLV_OPT)
 
 enum {
 	MLX5_HCA_CAP_OPMOD_GET_MAX   = 0,
@@ -1380,8 +1382,10 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 	u8 reserved_at_500[0x20];
 	u8 num_of_uars_per_page[0x20];
 	u8 flex_parser_protocols[0x20];
-	u8 reserved_at_560[0x20];
-	u8 reserved_at_580[0x3c];
+	u8 max_geneve_tlv_options[0x8];
+	u8 reserved_at_568[0x3];
+	u8 max_geneve_tlv_option_data_len[0x5];
+	u8 reserved_at_570[0x4c];
 	u8 mini_cqe_resp_stride_index[0x1];
 	u8 cqe_128_always[0x1];
 	u8 cqe_compression_128[0x1];
@@ -2292,6 +2296,7 @@ struct mlx5_ifc_create_cq_in_bits {
 };
 
 enum {
+	MLX5_OBJ_TYPE_GENEVE_TLV_OPT = 0x000b,
 	MLX5_GENERAL_OBJ_TYPE_VIRTQ = 0x000d,
 	MLX5_GENERAL_OBJ_TYPE_VIRTIO_Q_COUNTERS = 0x001c,
 	MLX5_GENERAL_OBJ_TYPE_FLEX_PARSE_GRAPH = 0x0022,
@@ -2326,6 +2331,17 @@ struct mlx5_ifc_virtio_q_counters_bits {
 	u8 reserved_at_180[0x50];
 };
 
+struct mlx5_ifc_geneve_tlv_option_bits {
+	u8 modify_field_select[0x40];
+	u8 reserved_at_40[0x18];
+	u8 geneve_option_fte_index[0x8];
+	u8 option_class[0x10];
+	u8 option_type[0x8];
+	u8 reserved_at_78[0x3];
+	u8 option_data_length[0x5];
+	u8 reserved_at_80[0x180];
+};
+
 struct mlx5_ifc_create_virtio_q_counters_in_bits {
 	struct mlx5_ifc_general_obj_in_cmd_hdr_bits hdr;
 	struct mlx5_ifc_virtio_q_counters_bits virtio_q_counters;
@@ -2335,6 +2351,12 @@ struct mlx5_ifc_query_virtio_q_counters_out_bits {
 	struct mlx5_ifc_general_obj_in_cmd_hdr_bits hdr;
 	struct mlx5_ifc_virtio_q_counters_bits virtio_q_counters;
 };
+
+struct mlx5_ifc_create_geneve_tlv_option_in_bits {
+	struct mlx5_ifc_general_obj_in_cmd_hdr_bits hdr;
+	struct mlx5_ifc_geneve_tlv_option_bits geneve_tlv_opt;
+};
+
 enum {
 	MLX5_VIRTQ_STATE_INIT = 0,
 	MLX5_VIRTQ_STATE_RDY = 1,
Signature · 1 line
2.21.0
#27 lore raw

PATCH v5 4/8 common/mlx5: create GENEVE TLV option object with DevX

In reply to Shiri Kuzin

TLV object is a special firmware maintained entity used to support match on GENEVE header extension option.

The TLV object is created with DevX API and accepts the option class, type and lehgth fields.

The class type and length fields are set using MLX5_SET and the Devx object is created using mlx5 glue function.

---
 drivers/common/mlx5/mlx5_devx_cmds.c          | 56 ++++++++++++++++++-
 drivers/common/mlx5/mlx5_devx_cmds.h          |  5 ++
 .../common/mlx5/rte_common_mlx5_exports.def   |  1 +
 drivers/common/mlx5/version.map               |  1 +
 4 files changed, 62 insertions(+), 1 deletion(-)
drivers/common/mlx5/mlx5_devx_cmds.c +55 -1
index ad54432eef..a5275f7ba1 100644
@@ -2071,7 +2071,6 @@ mlx5_devx_cmd_create_flow_hit_aso_obj(void *ctx, uint32_t pd)
  *
  * @param[in] ctx
  *   Context returned from mlx5 open_device() glue function.
- *
  * @return
  *   The DevX object created, NULL otherwise and rte_errno is set.
  */
@@ -2100,3 +2099,58 @@ mlx5_devx_cmd_alloc_pd(void *ctx)
 	ppd->id = MLX5_GET(alloc_pd_out, out, pd);
 	return ppd;
 }
+
+/**
+ * Create general object of type GENEVE TLV option using DevX API.
+ *
+ * @param[in] ctx
+ *   Context returned from mlx5 open_device() glue function.
+ * @param [in] class
+ *   TLV option variable value of class
+ * @param [in] type
+ *   TLV option variable value of type
+ * @param [in] len
+ *   TLV option variable value of len
+ *
+ * @return
+ *   The DevX object created, NULL otherwise and rte_errno is set.
+ */
+struct mlx5_devx_obj *
+mlx5_devx_cmd_create_geneve_tlv_option(void *ctx,
+		uint16_t class, uint8_t type, uint8_t len)
+{
+	uint32_t in[MLX5_ST_SZ_DW(create_geneve_tlv_option_in)] = {0};
+	uint32_t out[MLX5_ST_SZ_DW(general_obj_out_cmd_hdr)] = {0};
+	struct mlx5_devx_obj *geneve_tlv_opt_obj = mlx5_malloc(MLX5_MEM_ZERO,
+						   sizeof(*geneve_tlv_opt_obj),
+						   0, SOCKET_ID_ANY);
+
+	if (!geneve_tlv_opt_obj) {
+		DRV_LOG(ERR, "Failed to allocate geneve tlv option object.");
+		rte_errno = ENOMEM;
+		return NULL;
+	}
+	void *hdr = MLX5_ADDR_OF(create_geneve_tlv_option_in, in, hdr);
+	void *opt = MLX5_ADDR_OF(create_geneve_tlv_option_in, in,
+			geneve_tlv_opt);
+	MLX5_SET(general_obj_in_cmd_hdr, hdr, opcode,
+			MLX5_CMD_OP_CREATE_GENERAL_OBJECT);
+	MLX5_SET(general_obj_in_cmd_hdr, hdr, obj_type,
+			MLX5_OBJ_TYPE_GENEVE_TLV_OPT);
+	MLX5_SET(geneve_tlv_option, opt, option_class,
+			rte_be_to_cpu_16(class));
+	MLX5_SET(geneve_tlv_option, opt, option_type, type);
+	MLX5_SET(geneve_tlv_option, opt, option_data_length, len);
+	geneve_tlv_opt_obj->obj = mlx5_glue->devx_obj_create(ctx, in,
+					sizeof(in), out, sizeof(out));
+	if (!geneve_tlv_opt_obj->obj) {
+		rte_errno = errno;
+		DRV_LOG(ERR, "Failed to create Geneve tlv option "
+				"Obj using DevX.");
+		mlx5_free(geneve_tlv_opt_obj);
+		return NULL;
+	}
+	geneve_tlv_opt_obj->id = MLX5_GET(general_obj_out_cmd_hdr, out, obj_id);
+	return geneve_tlv_opt_obj;
+}
+
drivers/common/mlx5/mlx5_devx_cmds.h +5 -0
index 2286569f37..3a63ff0048 100644
@@ -486,6 +486,11 @@ __rte_internal
 int mlx5_devx_cmd_register_read(void *ctx, uint16_t reg_id,
 				uint32_t arg, uint32_t *data, uint32_t dw_cnt);
 
+__rte_internal
+struct mlx5_devx_obj *
+mlx5_devx_cmd_create_geneve_tlv_option(void *ctx,
+		uint16_t class, uint8_t type, uint8_t len);
+
 /**
  * Create virtio queue counters object DevX API.
  *
drivers/common/mlx5/rte_common_mlx5_exports.def +1 -0
index 9b7b7f4eae..1a7e54462d 100644
@@ -34,6 +34,7 @@ EXPORTS
 	mlx5_devx_cmd_register_read
 	mlx5_devx_get_out_command_status
 	mlx5_devx_cmd_create_flow_hit_aso_obj
+	mlx5_devx_cmd_create_geneve_tlv_option
 
 	mlx5_get_dbr
 	mlx5_glue
drivers/common/mlx5/version.map +1 -0
index fb3952bbc6..7de8006b89 100644
@@ -23,6 +23,7 @@ INTERNAL {
 	mlx5_devx_cmd_create_virtio_q_counters;
 	mlx5_devx_cmd_create_virtq;
         mlx5_devx_cmd_create_flow_hit_aso_obj;
+	mlx5_devx_cmd_create_geneve_tlv_option;
 	mlx5_devx_cmd_destroy;
 	mlx5_devx_cmd_flow_counter_alloc;
 	mlx5_devx_cmd_flow_counter_query;
Signature · 1 line
2.21.0
#28 lore raw

PATCH v5 5/8 net/mlx5: create GENEVE TLV option management

In reply to Shiri Kuzin

Currently firmware supports the only TLV object per device to match on the GENEVE header option.

This patch adds the simple TLV object management to the mlx5 PMD.

---
 drivers/net/mlx5/mlx5.c         |   2 +
 drivers/net/mlx5/mlx5.h         |  13 ++++
 drivers/net/mlx5/mlx5_flow.h    |   4 ++
 drivers/net/mlx5/mlx5_flow_dv.c | 108 ++++++++++++++++++++++++++++++++
 4 files changed, 127 insertions(+)
drivers/net/mlx5/mlx5.c +2 -0
index 023ef50a77..68d6352d48 100644
@@ -1014,6 +1014,7 @@ mlx5_alloc_shared_dev_ctx(const struct mlx5_dev_spawn_data *spawn,
 	rte_rwlock_write_unlock(&mlx5_shared_data->mem_event_rwlock);
 	/* Add context to the global device list. */
 	LIST_INSERT_HEAD(&mlx5_dev_ctx_list, sh, next);
+	rte_spinlock_init(&sh->geneve_tlv_opt_sl);
 exit:
 	pthread_mutex_unlock(&mlx5_dev_ctx_list_mutex);
 	return sh;
@@ -1109,6 +1110,7 @@ mlx5_free_shared_dev_ctx(struct mlx5_dev_ctx_shared *sh)
 		mlx5_glue->devx_free_uar(sh->devx_rx_uar);
 	if (sh->ctx)
 		claim_zero(mlx5_glue->close_device(sh->ctx));
+	MLX5_ASSERT(sh->geneve_tlv_option_resource == NULL);
 	pthread_mutex_destroy(&sh->txpp.mutex);
 	mlx5_free(sh);
 	return;
drivers/net/mlx5/mlx5.h +13 -0
index 41034f5d19..23272950a4 100644
@@ -540,6 +540,16 @@ struct mlx5_aso_age_mng {
 	struct mlx5_aso_sq aso_sq; /* ASO queue objects. */
 };
 
+/* Management structure for geneve tlv option */
+struct mlx5_geneve_tlv_option_resource {
+	struct mlx5_devx_obj *obj; /* Pointer to the geneve tlv opt object. */
+	rte_be16_t option_class; /* geneve tlv opt class.*/
+	uint8_t option_type; /* geneve tlv opt type.*/
+	uint8_t length; /* geneve tlv opt length. */
+	uint32_t refcnt; /* geneve tlv object reference counter */
+};
+
+
 #define MLX5_AGE_EVENT_NEW		1
 #define MLX5_AGE_TRIGGER		2
 #define MLX5_AGE_SET(age_info, BIT) \
@@ -752,6 +762,9 @@ struct mlx5_dev_ctx_shared {
 	void *devx_rx_uar; /* DevX UAR for Rx. */
 	struct mlx5_aso_age_mng *aso_age_mng;
 	/* Management data for aging mechanism using ASO Flow Hit. */
+	struct mlx5_geneve_tlv_option_resource *geneve_tlv_option_resource;
+	/* Management structure for geneve tlv option */
+	rte_spinlock_t geneve_tlv_opt_sl; /* Lock for geneve tlv resource */
 	struct mlx5_dev_shared_port port[]; /* per device port data array. */
 };
 
drivers/net/mlx5/mlx5_flow.h +4 -0
index ee85c9d8a5..0c8861964c 100644
@@ -1046,6 +1046,7 @@ struct rte_flow {
 	uint32_t counter; /**< Holds flow counter. */
 	uint32_t tunnel_id;  /**< Tunnel id */
 	uint32_t age; /**< Holds ASO age bit index. */
+	uint32_t geneve_tlv_option; /**< Holds Geneve TLV option id. > */
 } __rte_packed;
 
 /*
@@ -1503,4 +1504,7 @@ void flow_dv_dest_array_remove_cb(struct mlx5_cache_list *list,
 				  struct mlx5_cache_entry *entry);
 struct mlx5_aso_age_action *flow_aso_age_get_by_idx(struct rte_eth_dev *dev,
 						    uint32_t age_idx);
+int flow_dev_geneve_tlv_option_resource_register(struct rte_eth_dev *dev,
+					     const struct rte_flow_item *item,
+					     struct rte_flow_error *error);
 #endif /* RTE_PMD_MLX5_FLOW_H_ */
drivers/net/mlx5/mlx5_flow_dv.c +108 -0

126 lines not on this page · raw message

Signature · 1 line
2.21.0
#29 lore raw

PATCH v5 6/8 net/mlx5: add GENEVE TLV option flow validation

In reply to Shiri Kuzin

This patch adds validation routine for the GENEVE header TLV option.

The GENEVE TLV option match must include all fields with full masks due to NIC does not support masking on option class, type and length.

The option data length must be non zero and provided data pattern should be zero neither due to hardware limitations.

---
 drivers/net/mlx5/mlx5_flow.c    | 120 ++++++++++++++++++++++++++++++++
 drivers/net/mlx5/mlx5_flow.h    |   7 ++
 drivers/net/mlx5/mlx5_flow_dv.c |  10 ++-
 3 files changed, 136 insertions(+), 1 deletion(-)
drivers/net/mlx5/mlx5_flow.c +120 -0
index 2a4073c126..892ed2f2d9 100644
@@ -2627,6 +2627,126 @@ mlx5_flow_validate_item_geneve(const struct rte_flow_item *item,
 	return 0;
 }
 
+/**
+ * Validate Geneve TLV option item.
+ *
+ * @param[in] item
+ *   Item specification.
+ * @param[in] last_item
+ *   Previous validated item in the pattern items.
+ * @param[in] dev
+ *   Pointer to the rte_eth_dev structure.
+ * @param[out] error
+ *   Pointer to error structure.
+ *
+ * @return
+ *   0 on success, a negative errno value otherwise and rte_errno is set.
+ */
+int
+mlx5_flow_validate_item_geneve_opt(const struct rte_flow_item *item,
+				   uint64_t last_item,
+				   struct rte_eth_dev *dev,
+				   struct rte_flow_error *error)
+{
+	struct mlx5_priv *priv = dev->data->dev_private;
+	struct mlx5_dev_ctx_shared *sh = priv->sh;
+	struct mlx5_geneve_tlv_option_resource *geneve_opt_resource;
+	struct mlx5_hca_attr *hca_attr = &priv->config.hca_attr;
+	uint8_t data_max_supported =
+			hca_attr->max_geneve_tlv_option_data_len * 4;
+	struct mlx5_dev_config *config = &priv->config;
+	const struct rte_flow_item_geneve_opt *spec = item->spec;
+	const struct rte_flow_item_geneve_opt *mask = item->mask;
+	unsigned int i;
+	unsigned int data_len;
+	const struct rte_flow_item_geneve_opt full_mask = {
+		.option_class = RTE_BE16(0xffff),
+		.option_type = 0xff,
+		.option_len = 0x1f,
+	};
+
+	if (!mask)
+		mask = &rte_flow_item_geneve_opt_mask;
+	if (!spec)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt class/type/length must be specified");
+	if ((uint32_t)(spec->option_len) > MLX5_GENEVE_OPTLEN_MASK)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt length exceeeds the limit (31)");
+	/* Check if class type and length masks are full. */
+	if (full_mask.option_class != mask->option_class ||
+	    full_mask.option_type != mask->option_type ||
+	    full_mask.option_len != (mask->option_len & full_mask.option_len))
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt class/type/length masks must be full");
+	/* Check if length is supported */
+	if ((uint32_t)(spec->option_len) >
+			config->hca_attr.max_geneve_tlv_option_data_len)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt length not supported");
+	if (config->hca_attr.max_geneve_tlv_options > 1)
+		DRV_LOG(DEBUG,
+			"max_geneve_tlv_options supports more than 1 option");
+	/* Check GENEVE item preceding. */
+	if (!(last_item & MLX5_FLOW_LAYER_GENEVE))
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve opt item must be preceded with Geneve item");
+	/* Check if length is 0 or data is 0. */
+	if (spec->data == NULL || spec->option_len == 0)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt with zero data/length not supported");
+	/* Check not all data & mask are 0. */
+	data_len = spec->option_len * 4;
+	if (mask->data == NULL) {
+		for (i = 0; i < data_len; i++)
+			if (spec->data[i])
+				break;
+		if (i == data_len)
+			return rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_ITEM, item,
+				"Can't match on Geneve option data 0");
+	} else {
+		for (i = 0; i < data_len; i++)
+			if (spec->data[i] & mask->data[i])
+				break;
+		if (i == data_len)
+			return rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_ITEM, item,
+				"Can't match on Geneve option data and mask 0");
+		/* Check data mask supported. */
+		for (i = data_max_supported; i < data_len ; i++)
+			if (mask->data[i])
+				return rte_flow_error_set(error, ENOTSUP,
+					RTE_FLOW_ERROR_TYPE_ITEM, item,
+					"Data mask is of unsupported size");
+	}
+	/* Check GENEVE option is supported in NIC. */
+	if (!config->hca_attr.geneve_tlv_opt)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt not supported");
+	/* Check if we already have geneve option with different type/class. */
+	rte_spinlock_lock(&sh->geneve_tlv_opt_sl);
+	geneve_opt_resource = sh->geneve_tlv_option_resource;
+	if (geneve_opt_resource != NULL)
+		if (geneve_opt_resource->option_class != spec->option_class ||
+		    geneve_opt_resource->option_type != spec->option_type ||
+		    geneve_opt_resource->length != spec->option_len) {
+			rte_spinlock_unlock(&sh->geneve_tlv_opt_sl);
+			return rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_ITEM, item,
+				"Only one Geneve TLV option supported");
+		}
+	rte_spinlock_unlock(&sh->geneve_tlv_opt_sl);
+	return 0;
+}
+
 /**
  * Validate MPLS item.
  *
drivers/net/mlx5/mlx5_flow.h +7 -0
index 0c8861964c..28db5faee1 100644
@@ -138,6 +138,9 @@ enum mlx5_feature_name {
 #define MLX5_FLOW_LAYER_OUTER_L3_IPV6_FRAG_EXT (1u << 30)
 #define MLX5_FLOW_LAYER_INNER_L3_IPV6_FRAG_EXT (1u << 31)
 
+/* Pattern tunnel Layer bits (continued). */
+#define MLX5_FLOW_LAYER_GENEVE_OPT (UINT64_C(1) << 32)
+
 /* Outer Masks. */
 #define MLX5_FLOW_LAYER_OUTER_L3 \
 	(MLX5_FLOW_LAYER_OUTER_L3_IPV4 | MLX5_FLOW_LAYER_OUTER_L3_IPV6)
@@ -1396,6 +1399,10 @@ int mlx5_flow_validate_item_geneve(const struct rte_flow_item *item,
 				   uint64_t item_flags,
 				   struct rte_eth_dev *dev,
 				   struct rte_flow_error *error);
+int mlx5_flow_validate_item_geneve_opt(const struct rte_flow_item *item,
+				   uint64_t last_item,
+				   struct rte_eth_dev *dev,
+				   struct rte_flow_error *error);
 int mlx5_flow_validate_item_ecpri(const struct rte_flow_item *item,
 				  uint64_t item_flags,
 				  uint64_t last_item,
drivers/net/mlx5/mlx5_flow_dv.c +9 -1
index 0a657cea41..c24123fbf3 100644
@@ -5520,6 +5520,15 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr,
 				return ret;
 			last_item = MLX5_FLOW_LAYER_GENEVE;
 			break;
+		case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+			ret = mlx5_flow_validate_item_geneve_opt(items,
+								 last_item,
+								 dev,
+								 error);
+			if (ret < 0)
+				return ret;
+			last_item = MLX5_FLOW_LAYER_GENEVE_OPT;
+			break;
 		case RTE_FLOW_ITEM_TYPE_MPLS:
 			ret = mlx5_flow_validate_item_mpls(dev, items,
 							   item_flags,
@@ -5528,7 +5537,6 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr,
 				return ret;
 			last_item = MLX5_FLOW_LAYER_MPLS;
 			break;
-
 		case RTE_FLOW_ITEM_TYPE_MARK:
 			ret = flow_dv_validate_item_mark(dev, items, attr,
 							 error);
Signature · 1 line
2.21.0
#30 lore raw

PATCH v5 7/8 net/mlx5: add GENEVE TLV option flow translation

In reply to Shiri Kuzin

The GENEVE TLV option matching flows must be created using a translation function.

This function checks whether we already created a Devx object for the matching and either creates the objects or updates the reference counter.

---
 drivers/net/mlx5/mlx5_flow.h    |  2 +-
 drivers/net/mlx5/mlx5_flow_dv.c | 80 +++++++++++++++++++++++++++++++++
 2 files changed, 81 insertions(+), 1 deletion(-)
drivers/net/mlx5/mlx5_flow.h +1 -1
index 28db5faee1..d6a6839476 100644
@@ -329,7 +329,7 @@ enum mlx5_feature_name {
 #define MLX5_GENEVE_VER_VAL(a) \
 		(((a) >> (MLX5_GENEVE_VER_SHIFT)) & (MLX5_GENEVE_VER_MASK))
 #define MLX5_GENEVE_OPTLEN_MASK 0x3F
-#define MLX5_GENEVE_OPTLEN_SHIFT 7
+#define MLX5_GENEVE_OPTLEN_SHIFT 8
 #define MLX5_GENEVE_OPTLEN_VAL(a) \
 	    (((a) >> (MLX5_GENEVE_OPTLEN_SHIFT)) & (MLX5_GENEVE_OPTLEN_MASK))
 #define MLX5_GENEVE_OAMF_MASK 0x1
drivers/net/mlx5/mlx5_flow_dv.c +80 -0

92 lines not on this page · raw message

Signature · 1 line
2.21.0
#31 lore raw

PATCH v5 8/8 doc: update GENEVE TLV option support

In reply to Shiri Kuzin

GENEVE TLV option support added to mlx5 PMD.

The limitations and support were updated in
documentation.

---
 doc/guides/nics/mlx5.rst               | 23 ++++++++++++++++++++++-
 doc/guides/rel_notes/release_21_02.rst |  8 ++++++++
 2 files changed, 30 insertions(+), 1 deletion(-)
doc/guides/nics/mlx5.rst +22 -1
index 3bda0f8417..5f2b42b979 100644
@@ -98,6 +98,7 @@ Features
 - Hardware LRO.
 - Hairpin.
 - Multiple-thread flow insertion.
+- Matching on Geneve TLV option header with raw encap/decap action.
 
 Limitations
 -----------
@@ -175,7 +176,18 @@ Limitations
      - OAM
      - protocol type
      - options length
-       Currently, the only supported options length value is 0.
+
+- Match on Geneve TLv option is supported on the following fields:
+     - Class
+     - Type
+     - Length
+     - Data
+
+  Only one Class/Type/Length Geneve TLV option is supported per shared device.
+  Class/Type/Length fields must be specified as well as masks.
+  Class/Type/Length specified masks must be full.
+  Matching Geneve TLV option without specifying data is not supported.
+  Matching Geneve TLV option with data & mask == 0 is not supported.
 
 - VF: flow rules created on VF devices can only match traffic targeted at the
   configured MAC addresses (see ``rte_eth_dev_mac_addr_add()``).
@@ -1022,6 +1034,10 @@ Below are some firmware configurations listed.
    or
    FLEX_PARSER_PROFILE_ENABLE=1
 
+- enable Geneve TLV option flow matching::
+
+   FLEX_PARSER_PROFILE_ENABLE=0
+
 - enable GTP flow matching::
 
    FLEX_PARSER_PROFILE_ENABLE=3
@@ -1501,6 +1517,11 @@ Supported hardware offloads
    |                       | |  rdma-core 32 | | rdma-core 32  |
    |                       | |  ConnectX-6 Dx| | ConnectX-6 Dx |
    +-----------------------+-----------------+-----------------+
+   | Encapsulation         | | DPDK 21.02    | | DPDK 21.02    |
+   | GENEVE TLV option     | | OFED 5.2      | | OFED 5.2      |
+   |                       | | rdma-core 34  | | rdma-core 34  |
+   |                       | | ConnectX-6 Dx | | ConnectX-6 Dx |
+   +-----------------------+-----------------+-----------------+
 
 Notes for metadata
 ------------------
doc/guides/rel_notes/release_21_02.rst +8 -0
index 706cbf8f0c..1ac0232350 100644
@@ -55,6 +55,14 @@ New Features
      Also, make sure to start the actual text at the margin.
      =======================================================
 
+* **Updated Mellanox mlx5 driver.**
+
+  Updated the Mellanox mlx5 driver with new features and improvements, including:
+
+  * **Added GENEVE TLV option in rte_flow.**
+
+  Added support for matching GENEVE TLV option and raw encap/decap of GENEVE
+  TLV option.
 
 Removed Items
 -------------
Signature · 1 line
2.21.0
#32 lore raw

v6 cover ethdev: introduce GENEVE header TLV option item

In reply to Shiri Kuzin

The Geneve tunneling protocol is designed to allow the user to specify some data context on the packet.
The GENEVE TLV (Type-Length-Variable) Option is the mean intended to present the user data.

In order to support GENEVE TLV Option the new rte_flow item "rte_flow_item_geneve_opt" is introduced.
The new item contains the values and masks for the following fields:
-option class
-option type
-length
-data

The usage example:
"flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 / geneve-opt class is 5 length is 1 type is 0 data is 0x66998800 / end actions count / drop / end"

New item will be added to testpmd to support raw encap/decap action.

v6:
- added testpmd support for GENEVE option length.
- updated setting GENEVE option length value.

v5:
- added Devx function to def file.

v4:
- fixed setting option length.
- fixed testpmd calculation of GENEVE option size.
- updated documentation.

v3:
- updated documentation.

v2:
- removed pedantic.

Shiri Kuzin (7):
lib/librte_ethdev: introduce GENEVE header TLV option item common/mlx5: check GENEVE TLV support in HCA attributes common/mlx5: create GENEVE TLV option object with DevX net/mlx5: create GENEVE TLV option management net/mlx5: add GENEVE TLV option flow validation net/mlx5: add GENEVE TLV option flow translation
doc: update GENEVE TLV option support

Viacheslav Ovsiienko (2):
app/testpmd: add GENEVE option item support app/testpmd: add GENEVE header option length support

 app/test-pmd/cmdline_flow.c                   | 111 +++++++++-
 doc/guides/nics/mlx5.rst                      |  23 +-
 doc/guides/rel_notes/release_21_02.rst        |   8 +
 doc/guides/testpmd_app_ug/testpmd_funcs.rst   |   8 +
 drivers/common/mlx5/mlx5_devx_cmds.c          |  62 ++++++
 drivers/common/mlx5/mlx5_devx_cmds.h          |   9 +
 drivers/common/mlx5/mlx5_prm.h                |  28 ++-
 .../common/mlx5/rte_common_mlx5_exports.def   |   1 +
 drivers/common/mlx5/version.map               |   1 +
 drivers/net/mlx5/mlx5.c                       |   2 +
 drivers/net/mlx5/mlx5.h                       |  13 ++
 drivers/net/mlx5/mlx5_flow.c                  | 143 ++++++++++++
 drivers/net/mlx5/mlx5_flow.h                  |  15 +-
 drivers/net/mlx5/mlx5_flow_dv.c               | 205 ++++++++++++++++++
 lib/librte_ethdev/rte_flow.c                  |   1 +
 lib/librte_ethdev/rte_flow.h                  |  27 +++
 16 files changed, 646 insertions(+), 11 deletions(-)
Signature · 1 line
2.21.0
#33 lore raw

PATCH v6 1/9 lib/librte_ethdev: introduce GENEVE header TLV option item

The Geneve tunneling protocol is designed to allow the user to specify some data context on the packet. The GENEVE TLV (Type-Length-Variable) Option is the mean intended to present the user data.

In order to support GENEVE TLV Option the new rte_flow item "rte_flow_item_geneve_opt" is added.
The new item contains the values and masks for the following fields:
-option class
-option type
-length
-data

New item will be added to testpmd to support match and raw encap/decap actions.

---
 lib/librte_ethdev/rte_flow.c |  1 +
 lib/librte_ethdev/rte_flow.h | 27 +++++++++++++++++++++++++++
 2 files changed, 28 insertions(+)
lib/librte_ethdev/rte_flow.c +1 -0
index a06f64c271..2af7d965e1 100644
@@ -97,6 +97,7 @@ static const struct rte_flow_desc_data rte_flow_desc_item[] = {
 	MK_FLOW_ITEM(L2TPV3OIP, sizeof(struct rte_flow_item_l2tpv3oip)),
 	MK_FLOW_ITEM(PFCP, sizeof(struct rte_flow_item_pfcp)),
 	MK_FLOW_ITEM(ECPRI, sizeof(struct rte_flow_item_ecpri)),
+	MK_FLOW_ITEM(GENEVE_OPT, sizeof(struct rte_flow_item_geneve_opt)),
 };
 
 /** Generate flow_action[] entry. */
lib/librte_ethdev/rte_flow.h +27 -0
index 0977a78270..11a6494b8e 100644
@@ -543,6 +543,14 @@ enum rte_flow_item_type {
 	 * See struct rte_flow_item_ipv6_frag_ext.
 	 */
 	RTE_FLOW_ITEM_TYPE_IPV6_FRAG_EXT,
+
+	/**
+	 * Matches Geneve Variable Length Option
+	 *
+	 * See struct rte_flow_item_geneve_opt
+	 */
+	RTE_FLOW_ITEM_TYPE_GENEVE_OPT,
+
 };
 
 /**
@@ -1627,6 +1635,25 @@ static const struct rte_flow_item_ecpri rte_flow_item_ecpri_mask = {
 };
 #endif
 
+/**
+ * RTE_FLOW_ITEM_TYPE_GENEVE_OPT
+ *
+ * Matches a GENEVE Variable Length Option
+ */
+struct rte_flow_item_geneve_opt {
+	rte_be16_t option_class;
+	uint8_t option_type;
+	uint8_t option_len;
+	uint32_t *data;
+};
+
+/** Default mask for RTE_FLOW_ITEM_TYPE_GENEVE_OPT. */
+#ifndef __cplusplus
+static const struct rte_flow_item_geneve_opt
+rte_flow_item_geneve_opt_mask = {
+	.option_type = 0xff,
+};
+#endif
 /**
  * Matching pattern item definition.
  *
Signature · 1 line
2.21.0
#34 lore raw

PATCH v6 2/9 app/testpmd: add GENEVE option item support

In reply to Shiri Kuzin

From: Viacheslav Ovsiienko <[email protected]>

The patch adds the GENEVE option rte flow item support to command line interpreter. The flow command with GENEVE option items looks like:

flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 /
geneve-opt class is 99 length is 1 type is 0 data is 0x669988 / end actions drop / end

The option length should be specified in 32-bit words, this value specifies the length of the data pattern/mask arrays (should be multiplied by sizeof(uint32_t) to be expressed in bytes. If match on the length itself is not needed the mask should be set to zero, in this case length is used to specify the pattern/mask array lengths only.

---
 app/test-pmd/cmdline_flow.c                 | 101 ++++++++++++++++++--
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |   8 ++
 2 files changed, 103 insertions(+), 6 deletions(-)
app/test-pmd/cmdline_flow.c +95 -6
index 585cab98b4..98afde82cc 100644
@@ -283,6 +283,11 @@ enum index {
 	ITEM_ECPRI_MSG_IQ_DATA_PCID,
 	ITEM_ECPRI_MSG_RTC_CTRL_RTCID,
 	ITEM_ECPRI_MSG_DLY_MSR_MSRID,
+	ITEM_GENEVE_OPT,
+	ITEM_GENEVE_OPT_CLASS,
+	ITEM_GENEVE_OPT_TYPE,
+	ITEM_GENEVE_OPT_LENGTH,
+	ITEM_GENEVE_OPT_DATA,
 
 	/* Validate/create actions. */
 	ACTIONS,
@@ -413,6 +418,9 @@ enum index {
 /** Maximum size for pattern in struct rte_flow_item_raw. */
 #define ITEM_RAW_PATTERN_SIZE 40
 
+/** Maximum size for GENEVE option data pattern in bytes. */
+#define ITEM_GENEVE_OPT_DATA_SIZE 124
+
 /** Storage size for struct rte_flow_item_raw including pattern. */
 #define ITEM_RAW_SIZE \
 	(sizeof(struct rte_flow_item_raw) + ITEM_RAW_PATTERN_SIZE)
@@ -428,7 +436,7 @@ struct action_rss_data {
 };
 
 /** Maximum data size in struct rte_flow_action_raw_encap. */
-#define ACTION_RAW_ENCAP_MAX_DATA 128
+#define ACTION_RAW_ENCAP_MAX_DATA 512
 #define RAW_ENCAP_CONFS_MAX_NUM 8
 
 /** Storage for struct rte_flow_action_raw_encap. */
@@ -658,6 +666,16 @@ struct token {
 		.mask = (const void *)&(const s){ .f = (1 << (b)) - 1 }, \
 	})
 
+/** Static initializer for ARGS() to target a field with limits. */
+#define ARGS_ENTRY_BOUNDED(s, f, i, a) \
+	(&(const struct arg){ \
+		.bounded = 1, \
+		.min = (i), \
+		.max = (a), \
+		.offset = offsetof(s, f), \
+		.size = sizeof(((s *)0)->f), \
+	})
+
 /** Static initializer for ARGS() to target an arbitrary bit-mask. */
 #define ARGS_ENTRY_MASK(s, f, m) \
 	(&(const struct arg){ \
@@ -903,6 +921,7 @@ static const enum index next_item[] = {
 	ITEM_AH,
 	ITEM_PFCP,
 	ITEM_ECPRI,
+	ITEM_GENEVE_OPT,
 	END_SET,
 	ZERO,
 };
@@ -1244,6 +1263,15 @@ static const enum index item_ecpri_common_type[] = {
 	ZERO,
 };
 
+static const enum index item_geneve_opt[] = {
+	ITEM_GENEVE_OPT_CLASS,
+	ITEM_GENEVE_OPT_TYPE,
+	ITEM_GENEVE_OPT_LENGTH,
+	ITEM_GENEVE_OPT_DATA,
+	ITEM_NEXT,
+	ZERO,
+};
+
 static const enum index next_action[] = {
 	ACTION_END,
 	ACTION_VOID,
@@ -3230,6 +3258,47 @@ static const struct token token_list[] = {
 		.args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_ecpri,
 				hdr.type5.msr_id)),
 	},
+	[ITEM_GENEVE_OPT] = {
+		.name = "geneve-opt",
+		.help = "GENEVE header option",
+		.priv = PRIV_ITEM(GENEVE_OPT,
+				  sizeof(struct rte_flow_item_geneve_opt) +
+				  ITEM_GENEVE_OPT_DATA_SIZE),
+		.next = NEXT(item_geneve_opt),
+		.call = parse_vc,
+	},
+	[ITEM_GENEVE_OPT_CLASS]	= {
+		.name = "class",
+		.help = "GENEVE option class",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(UNSIGNED), item_param),
+		.args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_geneve_opt,
+					     option_class)),
+	},
+	[ITEM_GENEVE_OPT_TYPE] = {
+		.name = "type",
+		.help = "GENEVE option type",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(UNSIGNED), item_param),
+		.args = ARGS(ARGS_ENTRY(struct rte_flow_item_geneve_opt,
+					option_type)),
+	},
+	[ITEM_GENEVE_OPT_LENGTH] = {
+		.name = "length",
+		.help = "GENEVE option data length (in 32b words)",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(UNSIGNED), item_param),
+		.args = ARGS(ARGS_ENTRY_BOUNDED(
+				struct rte_flow_item_geneve_opt, option_len,
+				0, 31)),
+	},
+	[ITEM_GENEVE_OPT_DATA] = {
+		.name = "data",
+		.help = "GENEVE option data pattern",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(HEX), item_param),
+		.args = ARGS(ARGS_ENTRY(struct rte_flow_item_geneve_opt, data),
+			     ARGS_ENTRY_ARB(0, 0),
+			     ARGS_ENTRY_ARB
+				(sizeof(struct rte_flow_item_geneve_opt),
+				ITEM_GENEVE_OPT_DATA_SIZE)),
+	},
 	/* Validate/create actions. */
 	[ACTIONS] = {
 		.name = "actions",
@@ -6482,11 +6551,14 @@ parse_hex(struct context *ctx, const struct token *token,
 	ret = snprintf(tmp, sizeof(tmp), "%u", hexlen);
 	if (ret < 0)
 		goto error;
-	push_args(ctx, arg_len);
-	ret = parse_int(ctx, token, tmp, ret, NULL, 0);
-	if (ret < 0) {
-		pop_args(ctx);
-		goto error;
+	/* Save length if requested. */
+	if (arg_len->size) {
+		push_args(ctx, arg_len);
+		ret = parse_int(ctx, token, tmp, ret, NULL, 0);
+		if (ret < 0) {
+			pop_args(ctx);
+			goto error;
+		}
 	}
 	buf = (uint8_t *)ctx->object + arg_data->offset;
 	/* Output buffer is not necessarily NUL-terminated. */
@@ -7486,6 +7558,9 @@ flow_item_default_mask(const struct rte_flow_item *item)
 	case RTE_FLOW_ITEM_TYPE_GENEVE:
 		mask = &rte_flow_item_geneve_mask;
 		break;
+	case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+		mask = &rte_flow_item_geneve_opt_mask;
+		break;
 	case RTE_FLOW_ITEM_TYPE_PPPOE_PROTO_ID:
 		mask = &rte_flow_item_pppoe_proto_id_mask;
 		break;
@@ -7598,6 +7673,8 @@ cmd_set_raw_parsed(const struct buffer *in)
 	/* process hdr from upper layer to low layer (L3/L4 -> L2). */
 	data_tail = data + ACTION_RAW_ENCAP_MAX_DATA;
 	for (i = n - 1 ; i >= 0; --i) {
+		const struct rte_flow_item_geneve_opt *opt;
+
 		item = in->args.vc.pattern + i;
 		if (item->spec == NULL)
 			item->spec = flow_item_default_mask(item);
@@ -7650,6 +7727,18 @@ cmd_set_raw_parsed(const struct buffer *in)
 		case RTE_FLOW_ITEM_TYPE_GENEVE:
 			size = sizeof(struct rte_geneve_hdr);
 			break;
+		case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+			opt = (const struct rte_flow_item_geneve_opt *)
+								item->spec;
+			size = offsetof(struct rte_flow_item_geneve_opt, data);
+			if (opt->option_len && opt->data) {
+				*total_size += opt->option_len *
+					       sizeof(uint32_t);
+				rte_memcpy(data_tail - (*total_size),
+					   opt->data,
+					   opt->option_len * sizeof(uint32_t));
+			}
+			break;
 		case RTE_FLOW_ITEM_TYPE_L2TPV3OIP:
 			size = sizeof(rte_be32_t);
 			proto = 0x73;
doc/guides/testpmd_app_ug/testpmd_funcs.rst +8 -0
index 9be450066e..37278d31d6 100644
@@ -3680,6 +3680,14 @@ This section lists supported pattern items and their attributes, if any.
   - ``vni {unsigned}``: virtual network identifier.
   - ``protocol {unsigned}``: protocol type.
 
+- ``geneve-opt``: match GENEVE header option.
+
+  - ``class {unsigned}``: GENEVE option class.
+  - ``type {unsigned}``: GENEVE option type.
+  - ``length {unsigned}``: GENEVE option length in 32-bit words.
+  - ``data {hex string}``: GENEVE option data, the legnt is defined by
+    ``length`` field
+
 - ``vxlan-gpe``: match VXLAN-GPE header.
 
   - ``vni {unsigned}``: VXLAN-GPE identifier.
Signature · 1 line
2.21.0
#35 lore raw

PATCH v6 3/9 app/testpmd: add GENEVE header option length support

In reply to Shiri Kuzin

From: Viacheslav Ovsiienko <[email protected]>

The patch adds the GENEVE rte flow option length support to command line interpreter. The flow command with GENEVE option items looks like:

flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 optlen is 2 / end actions drop / end

The option length should be specified in 32-bit words, this value specifies the all options length in the GENEVE header.

---
 app/test-pmd/cmdline_flow.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
app/test-pmd/cmdline_flow.c +10 -0
index 98afde82cc..1ef4d4f75d 100644
@@ -223,6 +223,7 @@ enum index {
 	ITEM_GENEVE,
 	ITEM_GENEVE_VNI,
 	ITEM_GENEVE_PROTO,
+	ITEM_GENEVE_OPTLEN,
 	ITEM_VXLAN_GPE,
 	ITEM_VXLAN_GPE_VNI,
 	ITEM_ARP_ETH_IPV4,
@@ -1101,6 +1102,7 @@ static const enum index item_gtp[] = {
 static const enum index item_geneve[] = {
 	ITEM_GENEVE_VNI,
 	ITEM_GENEVE_PROTO,
+	ITEM_GENEVE_OPTLEN,
 	ITEM_NEXT,
 	ZERO,
 };
@@ -2807,6 +2809,14 @@ static const struct token token_list[] = {
 		.args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_geneve,
 					     protocol)),
 	},
+	[ITEM_GENEVE_OPTLEN] = {
+		.name = "optlen",
+		.help = "GENEVE options length in dwords",
+		.next = NEXT(item_geneve, NEXT_ENTRY(UNSIGNED), item_param),
+		.args = ARGS(ARGS_ENTRY_MASK_HTON(struct rte_flow_item_geneve,
+						  ver_opt_len_o_c_rsvd0,
+						  "\x3f\x00")),
+	},
 	[ITEM_VXLAN_GPE] = {
 		.name = "vxlan-gpe",
 		.help = "match VXLAN-GPE header",
Signature · 1 line
2.21.0
Acked-by on v6 3/9
#36 lore raw

Re: [dpdk-dev] [PATCH v6 3/9] app/testpmd: add GENEVE header option length support

Quoted 65 lines

-----Original Message-----
From: Shiri Kuzin <[email protected]>
Sent: Thursday, January 14, 2021 9:08 AM
To: [email protected]
Subject: [PATCH v6 3/9] app/testpmd: add GENEVE header option length support

From: Viacheslav Ovsiienko <[email protected]>

The patch adds the GENEVE rte flow option length support to command line interpreter. The flow command with GENEVE option items looks like:

flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 optlen is 2 / end actions drop / end

The option length should be specified in 32-bit words, this value specifies the all options length in the GENEVE header.

---
 app/test-pmd/cmdline_flow.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
app/test-pmd/cmdline_flow.c +10 -0
index 98afde82cc..1ef4d4f75d 100644
@@ -223,6 +223,7 @@ enum index {
 	ITEM_GENEVE,
 	ITEM_GENEVE_VNI,
 	ITEM_GENEVE_PROTO,
+	ITEM_GENEVE_OPTLEN,
 	ITEM_VXLAN_GPE,
 	ITEM_VXLAN_GPE_VNI,
 	ITEM_ARP_ETH_IPV4,
@@ -1101,6 +1102,7 @@ static const enum index item_gtp[] = {
 static const enum index item_geneve[] = {
 	ITEM_GENEVE_VNI,
 	ITEM_GENEVE_PROTO,
+	ITEM_GENEVE_OPTLEN,
 	ITEM_NEXT,
 	ZERO,
 };
@@ -2807,6 +2809,14 @@ static const struct token token_list[] = {
 		.args = ARGS(ARGS_ENTRY_HTON(struct
 rte_flow_item_geneve,
 					     protocol)),
 	},
+	[ITEM_GENEVE_OPTLEN] = {
+		.name = "optlen",
+		.help = "GENEVE options length in dwords",
+		.next = NEXT(item_geneve, NEXT_ENTRY(UNSIGNED),
 item_param),
+		.args = ARGS(ARGS_ENTRY_MASK_HTON(struct
 rte_flow_item_geneve,
+						  ver_opt_len_o_c_rsvd0,
+						  "\x3f\x00")),
+	},
app/test-pmd/cmdline_flow.c +0 -1
 	[ITEM_VXLAN_GPE] = {
 		.name = "vxlan-gpe",
 		.help = "match VXLAN-GPE header",
--
Quoted 1 line

2.21.0

Thanks,
Ori

#37 lore raw

PATCH v6 4/9 common/mlx5: check GENEVE TLV support in HCA attributes

In reply to Shiri Kuzin

This is preparation step to support match on GENEVE TLV option.

In this Patch we add the HCA attributes that will allow supporting GENEVE TLV option matching.

---
 drivers/common/mlx5/mlx5_devx_cmds.c |  7 +++++++
 drivers/common/mlx5/mlx5_devx_cmds.h |  4 ++++
 drivers/common/mlx5/mlx5_prm.h       | 28 +++++++++++++++++++++++++---
 3 files changed, 36 insertions(+), 3 deletions(-)
drivers/common/mlx5/mlx5_devx_cmds.c +7 -0
index 4d01f52986..861f75e47d 100644
@@ -690,6 +690,10 @@ mlx5_devx_cmd_query_hca_attr(void *ctx,
 	attr->eth_virt = MLX5_GET(cmd_hca_cap, hcattr, eth_virt);
 	attr->flex_parser_protocols = MLX5_GET(cmd_hca_cap, hcattr,
 					       flex_parser_protocols);
+	attr->max_geneve_tlv_options = MLX5_GET(cmd_hca_cap, hcattr,
+			max_geneve_tlv_options);
+	attr->max_geneve_tlv_option_data_len = MLX5_GET(cmd_hca_cap, hcattr,
+			max_geneve_tlv_option_data_len);
 	attr->qos.sup = MLX5_GET(cmd_hca_cap, hcattr, qos);
 	attr->vdpa.valid = !!(MLX5_GET64(cmd_hca_cap, hcattr,
 					 general_obj_types) &
@@ -717,6 +721,9 @@ mlx5_devx_cmd_query_hca_attr(void *ctx,
 	attr->flow_hit_aso = !!(MLX5_GET64(cmd_hca_cap, hcattr,
 					   general_obj_types) &
 				MLX5_GENERAL_OBJ_TYPES_CAP_FLOW_HIT_ASO);
+	attr->geneve_tlv_opt = !!(MLX5_GET64(cmd_hca_cap, hcattr,
+					   general_obj_types) &
+				MLX5_GENERAL_OBJ_TYPES_CAP_GENEVE_TLV_OPT);
 	attr->log_max_cq = MLX5_GET(cmd_hca_cap, hcattr, log_max_cq);
 	attr->log_max_qp = MLX5_GET(cmd_hca_cap, hcattr, log_max_qp);
 	attr->log_max_cq_sz = MLX5_GET(cmd_hca_cap, hcattr, log_max_cq_sz);
drivers/common/mlx5/mlx5_devx_cmds.h +4 -0
index 8d993dfad7..1e0a48d810 100644
@@ -97,6 +97,8 @@ struct mlx5_hca_attr {
 	uint32_t lro_timer_supported_periods[MLX5_LRO_NUM_SUPP_PERIODS];
 	uint16_t lro_min_mss_size;
 	uint32_t flex_parser_protocols;
+	uint32_t max_geneve_tlv_options;
+	uint32_t max_geneve_tlv_option_data_len;
 	uint32_t hairpin:1;
 	uint32_t log_max_hairpin_queues:5;
 	uint32_t log_max_hairpin_wq_data_sz:5;
@@ -116,6 +118,7 @@ struct mlx5_hca_attr {
 	uint32_t regex:1;
 	uint32_t regexp_num_of_engines;
 	uint32_t log_max_ft_sampler_num:8;
+	uint32_t geneve_tlv_opt;
 	struct mlx5_hca_qos_attr qos;
 	struct mlx5_hca_vdpa_attr vdpa;
 	int log_max_qp_sz;
@@ -481,6 +484,7 @@ struct mlx5_devx_obj *mlx5_devx_cmd_create_flex_parser(void *ctx,
 __rte_internal
 int mlx5_devx_cmd_register_read(void *ctx, uint16_t reg_id,
 				uint32_t arg, uint32_t *data, uint32_t dw_cnt);
+
 /**
  * Create virtio queue counters object DevX API.
  *
drivers/common/mlx5/mlx5_prm.h +25 -3
index 9e2d1d0c81..e37c8f9a3c 100644
@@ -789,7 +789,7 @@ struct mlx5_ifc_fte_match_set_misc3_bits {
 	u8 icmp_code[0x8];
 	u8 icmpv6_type[0x8];
 	u8 icmpv6_code[0x8];
-	u8 reserved_at_120[0x20];
+	u8 geneve_tlv_option_0_data[0x20];
 	u8 gtpu_teid[0x20];
 	u8 gtpu_msg_type[0x08];
 	u8 gtpu_msg_flags[0x08];
@@ -1082,6 +1082,8 @@ enum {
 			(1ULL << MLX5_GENERAL_OBJ_TYPE_FLEX_PARSE_GRAPH)
 #define MLX5_GENERAL_OBJ_TYPES_CAP_FLOW_HIT_ASO \
 			(1ULL << MLX5_GENERAL_OBJ_TYPE_FLOW_HIT_ASO)
+#define MLX5_GENERAL_OBJ_TYPES_CAP_GENEVE_TLV_OPT \
+			(1ULL << MLX5_OBJ_TYPE_GENEVE_TLV_OPT)
 
 enum {
 	MLX5_HCA_CAP_OPMOD_GET_MAX   = 0,
@@ -1380,8 +1382,10 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 	u8 reserved_at_500[0x20];
 	u8 num_of_uars_per_page[0x20];
 	u8 flex_parser_protocols[0x20];
-	u8 reserved_at_560[0x20];
-	u8 reserved_at_580[0x3c];
+	u8 max_geneve_tlv_options[0x8];
+	u8 reserved_at_568[0x3];
+	u8 max_geneve_tlv_option_data_len[0x5];
+	u8 reserved_at_570[0x4c];
 	u8 mini_cqe_resp_stride_index[0x1];
 	u8 cqe_128_always[0x1];
 	u8 cqe_compression_128[0x1];
@@ -2292,6 +2296,7 @@ struct mlx5_ifc_create_cq_in_bits {
 };
 
 enum {
+	MLX5_OBJ_TYPE_GENEVE_TLV_OPT = 0x000b,
 	MLX5_GENERAL_OBJ_TYPE_VIRTQ = 0x000d,
 	MLX5_GENERAL_OBJ_TYPE_VIRTIO_Q_COUNTERS = 0x001c,
 	MLX5_GENERAL_OBJ_TYPE_FLEX_PARSE_GRAPH = 0x0022,
@@ -2326,6 +2331,17 @@ struct mlx5_ifc_virtio_q_counters_bits {
 	u8 reserved_at_180[0x50];
 };
 
+struct mlx5_ifc_geneve_tlv_option_bits {
+	u8 modify_field_select[0x40];
+	u8 reserved_at_40[0x18];
+	u8 geneve_option_fte_index[0x8];
+	u8 option_class[0x10];
+	u8 option_type[0x8];
+	u8 reserved_at_78[0x3];
+	u8 option_data_length[0x5];
+	u8 reserved_at_80[0x180];
+};
+
 struct mlx5_ifc_create_virtio_q_counters_in_bits {
 	struct mlx5_ifc_general_obj_in_cmd_hdr_bits hdr;
 	struct mlx5_ifc_virtio_q_counters_bits virtio_q_counters;
@@ -2335,6 +2351,12 @@ struct mlx5_ifc_query_virtio_q_counters_out_bits {
 	struct mlx5_ifc_general_obj_in_cmd_hdr_bits hdr;
 	struct mlx5_ifc_virtio_q_counters_bits virtio_q_counters;
 };
+
+struct mlx5_ifc_create_geneve_tlv_option_in_bits {
+	struct mlx5_ifc_general_obj_in_cmd_hdr_bits hdr;
+	struct mlx5_ifc_geneve_tlv_option_bits geneve_tlv_opt;
+};
+
 enum {
 	MLX5_VIRTQ_STATE_INIT = 0,
 	MLX5_VIRTQ_STATE_RDY = 1,
Signature · 1 line
2.21.0
#38 lore raw

PATCH v6 5/9 common/mlx5: create GENEVE TLV option object with DevX

In reply to Shiri Kuzin

TLV object is a special firmware maintained entity used to support match on GENEVE header extension option.

The TLV object is created with DevX API and accepts the option class, type and lehgth fields.

The class type and length fields are set using MLX5_SET and the Devx object is created using mlx5 glue function.

---
 drivers/common/mlx5/mlx5_devx_cmds.c          | 55 +++++++++++++++++++
 drivers/common/mlx5/mlx5_devx_cmds.h          |  5 ++
 .../common/mlx5/rte_common_mlx5_exports.def   |  1 +
 drivers/common/mlx5/version.map               |  1 +
 4 files changed, 62 insertions(+)
drivers/common/mlx5/mlx5_devx_cmds.c +55 -0
index 861f75e47d..d5859c2a26 100644
@@ -2093,3 +2093,58 @@ mlx5_devx_cmd_alloc_pd(void *ctx)
 	ppd->id = MLX5_GET(alloc_pd_out, out, pd);
 	return ppd;
 }
+
+/**
+ * Create general object of type GENEVE TLV option using DevX API.
+ *
+ * @param[in] ctx
+ *   Context returned from mlx5 open_device() glue function.
+ * @param [in] class
+ *   TLV option variable value of class
+ * @param [in] type
+ *   TLV option variable value of type
+ * @param [in] len
+ *   TLV option variable value of len
+ *
+ * @return
+ *   The DevX object created, NULL otherwise and rte_errno is set.
+ */
+struct mlx5_devx_obj *
+mlx5_devx_cmd_create_geneve_tlv_option(void *ctx,
+		uint16_t class, uint8_t type, uint8_t len)
+{
+	uint32_t in[MLX5_ST_SZ_DW(create_geneve_tlv_option_in)] = {0};
+	uint32_t out[MLX5_ST_SZ_DW(general_obj_out_cmd_hdr)] = {0};
+	struct mlx5_devx_obj *geneve_tlv_opt_obj = mlx5_malloc(MLX5_MEM_ZERO,
+						   sizeof(*geneve_tlv_opt_obj),
+						   0, SOCKET_ID_ANY);
+
+	if (!geneve_tlv_opt_obj) {
+		DRV_LOG(ERR, "Failed to allocate geneve tlv option object.");
+		rte_errno = ENOMEM;
+		return NULL;
+	}
+	void *hdr = MLX5_ADDR_OF(create_geneve_tlv_option_in, in, hdr);
+	void *opt = MLX5_ADDR_OF(create_geneve_tlv_option_in, in,
+			geneve_tlv_opt);
+	MLX5_SET(general_obj_in_cmd_hdr, hdr, opcode,
+			MLX5_CMD_OP_CREATE_GENERAL_OBJECT);
+	MLX5_SET(general_obj_in_cmd_hdr, hdr, obj_type,
+			MLX5_OBJ_TYPE_GENEVE_TLV_OPT);
+	MLX5_SET(geneve_tlv_option, opt, option_class,
+			rte_be_to_cpu_16(class));
+	MLX5_SET(geneve_tlv_option, opt, option_type, type);
+	MLX5_SET(geneve_tlv_option, opt, option_data_length, len);
+	geneve_tlv_opt_obj->obj = mlx5_glue->devx_obj_create(ctx, in,
+					sizeof(in), out, sizeof(out));
+	if (!geneve_tlv_opt_obj->obj) {
+		rte_errno = errno;
+		DRV_LOG(ERR, "Failed to create Geneve tlv option "
+				"Obj using DevX.");
+		mlx5_free(geneve_tlv_opt_obj);
+		return NULL;
+	}
+	geneve_tlv_opt_obj->id = MLX5_GET(general_obj_out_cmd_hdr, out, obj_id);
+	return geneve_tlv_opt_obj;
+}
+
drivers/common/mlx5/mlx5_devx_cmds.h +5 -0
index 1e0a48d810..bf83a903b1 100644
@@ -485,6 +485,11 @@ __rte_internal
 int mlx5_devx_cmd_register_read(void *ctx, uint16_t reg_id,
 				uint32_t arg, uint32_t *data, uint32_t dw_cnt);
 
+__rte_internal
+struct mlx5_devx_obj *
+mlx5_devx_cmd_create_geneve_tlv_option(void *ctx,
+		uint16_t class, uint8_t type, uint8_t len);
+
 /**
  * Create virtio queue counters object DevX API.
  *
drivers/common/mlx5/rte_common_mlx5_exports.def +1 -0
index b385d38d1a..fd62b806ca 100644
@@ -34,6 +34,7 @@ EXPORTS
 	mlx5_devx_cmd_register_read
 	mlx5_devx_get_out_command_status
 	mlx5_devx_cmd_create_flow_hit_aso_obj
+	mlx5_devx_cmd_create_geneve_tlv_option
 
 	mlx5_devx_cq_create
 	mlx5_devx_cq_destroy
drivers/common/mlx5/version.map +1 -0
index 00760abece..244b9c7339 100644
@@ -23,6 +23,7 @@ INTERNAL {
 	mlx5_devx_cmd_create_virtio_q_counters;
 	mlx5_devx_cmd_create_virtq;
         mlx5_devx_cmd_create_flow_hit_aso_obj;
+	mlx5_devx_cmd_create_geneve_tlv_option;
 	mlx5_devx_cmd_destroy;
 	mlx5_devx_cmd_flow_counter_alloc;
 	mlx5_devx_cmd_flow_counter_query;
Signature · 1 line
2.21.0
#39 lore raw

PATCH v6 6/9 net/mlx5: create GENEVE TLV option management

In reply to Shiri Kuzin

Currently firmware supports the only TLV object per device to match on the GENEVE header option.

This patch adds the simple TLV object management to the mlx5 PMD.

---
 drivers/net/mlx5/mlx5.c         |   2 +
 drivers/net/mlx5/mlx5.h         |  13 ++++
 drivers/net/mlx5/mlx5_flow.h    |   5 ++
 drivers/net/mlx5/mlx5_flow_dv.c | 108 ++++++++++++++++++++++++++++++++
 4 files changed, 128 insertions(+)
drivers/net/mlx5/mlx5.c +2 -0
index e245276fce..3730f32295 100644
@@ -1008,6 +1008,7 @@ mlx5_alloc_shared_dev_ctx(const struct mlx5_dev_spawn_data *spawn,
 	rte_rwlock_write_unlock(&mlx5_shared_data->mem_event_rwlock);
 	/* Add context to the global device list. */
 	LIST_INSERT_HEAD(&mlx5_dev_ctx_list, sh, next);
+	rte_spinlock_init(&sh->geneve_tlv_opt_sl);
 exit:
 	pthread_mutex_unlock(&mlx5_dev_ctx_list_mutex);
 	return sh;
@@ -1106,6 +1107,7 @@ mlx5_free_shared_dev_ctx(struct mlx5_dev_ctx_shared *sh)
 		mlx5_glue->devx_free_uar(sh->devx_rx_uar);
 	if (sh->ctx)
 		claim_zero(mlx5_glue->close_device(sh->ctx));
+	MLX5_ASSERT(sh->geneve_tlv_option_resource == NULL);
 	pthread_mutex_destroy(&sh->txpp.mutex);
 	mlx5_free(sh);
 	return;
drivers/net/mlx5/mlx5.h +13 -0
index 3836a9696c..101e9c20d0 100644
@@ -528,6 +528,16 @@ struct mlx5_aso_age_mng {
 	struct mlx5_aso_sq aso_sq; /* ASO queue objects. */
 };
 
+/* Management structure for geneve tlv option */
+struct mlx5_geneve_tlv_option_resource {
+	struct mlx5_devx_obj *obj; /* Pointer to the geneve tlv opt object. */
+	rte_be16_t option_class; /* geneve tlv opt class.*/
+	uint8_t option_type; /* geneve tlv opt type.*/
+	uint8_t length; /* geneve tlv opt length. */
+	uint32_t refcnt; /* geneve tlv object reference counter */
+};
+
+
 #define MLX5_AGE_EVENT_NEW		1
 #define MLX5_AGE_TRIGGER		2
 #define MLX5_AGE_SET(age_info, BIT) \
@@ -727,6 +737,9 @@ struct mlx5_dev_ctx_shared {
 	void *devx_rx_uar; /* DevX UAR for Rx. */
 	struct mlx5_aso_age_mng *aso_age_mng;
 	/* Management data for aging mechanism using ASO Flow Hit. */
+	struct mlx5_geneve_tlv_option_resource *geneve_tlv_option_resource;
+	/* Management structure for geneve tlv option */
+	rte_spinlock_t geneve_tlv_opt_sl; /* Lock for geneve tlv resource */
 	struct mlx5_dev_shared_port port[]; /* per device port data array. */
 };
 
drivers/net/mlx5/mlx5_flow.h +5 -0
index 329082a48b..3d6ba9f610 100644
@@ -1046,6 +1046,7 @@ struct rte_flow {
 	uint32_t counter; /**< Holds flow counter. */
 	uint32_t tunnel_id;  /**< Tunnel id */
 	uint32_t age; /**< Holds ASO age bit index. */
+	uint32_t geneve_tlv_option; /**< Holds Geneve TLV option id. > */
 } __rte_packed;
 
 /*
@@ -1503,6 +1504,9 @@ void flow_dv_dest_array_remove_cb(struct mlx5_cache_list *list,
 				  struct mlx5_cache_entry *entry);
 struct mlx5_aso_age_action *flow_aso_age_get_by_idx(struct rte_eth_dev *dev,
 						    uint32_t age_idx);
+int flow_dev_geneve_tlv_option_resource_register(struct rte_eth_dev *dev,
+					     const struct rte_flow_item *item,
+					     struct rte_flow_error *error);
 
 void flow_release_workspace(void *data);
 int mlx5_flow_os_init_workspace_once(void);
@@ -1510,4 +1514,5 @@ void *mlx5_flow_os_get_specific_workspace(void);
 int mlx5_flow_os_set_specific_workspace(struct mlx5_flow_workspace *data);
 void mlx5_flow_os_release_workspace(void);
 
+
 #endif /* RTE_PMD_MLX5_FLOW_H_ */
drivers/net/mlx5/mlx5_flow_dv.c +108 -0

126 lines not on this page · raw message

Signature · 1 line
2.21.0
#40 lore raw

PATCH v6 7/9 net/mlx5: add GENEVE TLV option flow validation

In reply to Shiri Kuzin

This patch adds validation routine for the GENEVE header TLV option.

The GENEVE TLV option match must include all fields with full masks due to NIC does not support masking on option class, type and length.

The option data length must be non zero and provided data pattern should be zero neither due to hardware limitations.

---
 drivers/net/mlx5/mlx5_flow.c    | 143 ++++++++++++++++++++++++++++++++
 drivers/net/mlx5/mlx5_flow.h    |   8 ++
 drivers/net/mlx5/mlx5_flow_dv.c |  12 +++
 3 files changed, 163 insertions(+)
drivers/net/mlx5/mlx5_flow.c +143 -0
index 1790e774ba..3c4ef6d7d2 100644
@@ -2622,6 +2622,149 @@ mlx5_flow_validate_item_geneve(const struct rte_flow_item *item,
 	return 0;
 }
 
+/**
+ * Validate Geneve TLV option item.
+ *
+ * @param[in] item
+ *   Item specification.
+ * @param[in] last_item
+ *   Previous validated item in the pattern items.
+ * @param[in] geneve_item
+ *   Previous GENEVE item specification.
+ * @param[in] dev
+ *   Pointer to the rte_eth_dev structure.
+ * @param[out] error
+ *   Pointer to error structure.
+ *
+ * @return
+ *   0 on success, a negative errno value otherwise and rte_errno is set.
+ */
+int
+mlx5_flow_validate_item_geneve_opt(const struct rte_flow_item *item,
+				   uint64_t last_item,
+				   const struct rte_flow_item *geneve_item,
+				   struct rte_eth_dev *dev,
+				   struct rte_flow_error *error)
+{
+	struct mlx5_priv *priv = dev->data->dev_private;
+	struct mlx5_dev_ctx_shared *sh = priv->sh;
+	struct mlx5_geneve_tlv_option_resource *geneve_opt_resource;
+	struct mlx5_hca_attr *hca_attr = &priv->config.hca_attr;
+	uint8_t data_max_supported =
+			hca_attr->max_geneve_tlv_option_data_len * 4;
+	struct mlx5_dev_config *config = &priv->config;
+	const struct rte_flow_item_geneve *geneve_spec;
+	const struct rte_flow_item_geneve *geneve_mask;
+	const struct rte_flow_item_geneve_opt *spec = item->spec;
+	const struct rte_flow_item_geneve_opt *mask = item->mask;
+	unsigned int i;
+	unsigned int data_len;
+	uint8_t tlv_option_len;
+	uint16_t optlen_m, optlen_v;
+	const struct rte_flow_item_geneve_opt full_mask = {
+		.option_class = RTE_BE16(0xffff),
+		.option_type = 0xff,
+		.option_len = 0x1f,
+	};
+
+	if (!mask)
+		mask = &rte_flow_item_geneve_opt_mask;
+	if (!spec)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt class/type/length must be specified");
+	if ((uint32_t)spec->option_len > MLX5_GENEVE_OPTLEN_MASK)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt length exceeeds the limit (31)");
+	/* Check if class type and length masks are full. */
+	if (full_mask.option_class != mask->option_class ||
+	    full_mask.option_type != mask->option_type ||
+	    full_mask.option_len != (mask->option_len & full_mask.option_len))
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt class/type/length masks must be full");
+	/* Check if length is supported */
+	if ((uint32_t)spec->option_len >
+			config->hca_attr.max_geneve_tlv_option_data_len)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt length not supported");
+	if (config->hca_attr.max_geneve_tlv_options > 1)
+		DRV_LOG(DEBUG,
+			"max_geneve_tlv_options supports more than 1 option");
+	/* Check GENEVE item preceding. */
+	if (!geneve_item || !(last_item & MLX5_FLOW_LAYER_GENEVE))
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve opt item must be preceded with Geneve item");
+	geneve_spec = geneve_item->spec;
+	geneve_mask = geneve_item->mask ? geneve_item->mask :
+					  &rte_flow_item_geneve_mask;
+	/* Check if GENEVE TLV option size doesn't exceed option length */
+	if (geneve_spec && (geneve_mask->ver_opt_len_o_c_rsvd0 ||
+			    geneve_spec->ver_opt_len_o_c_rsvd0)) {
+		tlv_option_len = spec->option_len & mask->option_len;
+		optlen_v = rte_be_to_cpu_16(geneve_spec->ver_opt_len_o_c_rsvd0);
+		optlen_v = MLX5_GENEVE_OPTLEN_VAL(optlen_v);
+		optlen_m = rte_be_to_cpu_16(geneve_mask->ver_opt_len_o_c_rsvd0);
+		optlen_m = MLX5_GENEVE_OPTLEN_VAL(optlen_m);
+		if ((optlen_v & optlen_m) <= tlv_option_len)
+			return rte_flow_error_set
+				(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+				 "GENEVE TLV option length exceeds optlen");
+	}
+	/* Check if length is 0 or data is 0. */
+	if (spec->data == NULL || spec->option_len == 0)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt with zero data/length not supported");
+	/* Check not all data & mask are 0. */
+	data_len = spec->option_len * 4;
+	if (mask->data == NULL) {
+		for (i = 0; i < data_len; i++)
+			if (spec->data[i])
+				break;
+		if (i == data_len)
+			return rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_ITEM, item,
+				"Can't match on Geneve option data 0");
+	} else {
+		for (i = 0; i < data_len; i++)
+			if (spec->data[i] & mask->data[i])
+				break;
+		if (i == data_len)
+			return rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_ITEM, item,
+				"Can't match on Geneve option data and mask 0");
+		/* Check data mask supported. */
+		for (i = data_max_supported; i < data_len ; i++)
+			if (mask->data[i])
+				return rte_flow_error_set(error, ENOTSUP,
+					RTE_FLOW_ERROR_TYPE_ITEM, item,
+					"Data mask is of unsupported size");
+	}
+	/* Check GENEVE option is supported in NIC. */
+	if (!config->hca_attr.geneve_tlv_opt)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt not supported");
+	/* Check if we already have geneve option with different type/class. */
+	rte_spinlock_lock(&sh->geneve_tlv_opt_sl);
+	geneve_opt_resource = sh->geneve_tlv_option_resource;
+	if (geneve_opt_resource != NULL)
+		if (geneve_opt_resource->option_class != spec->option_class ||
+		    geneve_opt_resource->option_type != spec->option_type ||
+		    geneve_opt_resource->length != spec->option_len) {
+			rte_spinlock_unlock(&sh->geneve_tlv_opt_sl);
+			return rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_ITEM, item,
+				"Only one Geneve TLV option supported");
+		}
+	rte_spinlock_unlock(&sh->geneve_tlv_opt_sl);
+	return 0;
+}
+
 /**
  * Validate MPLS item.
  *
drivers/net/mlx5/mlx5_flow.h +8 -0
index 3d6ba9f610..e52f902448 100644
@@ -138,6 +138,9 @@ enum mlx5_feature_name {
 #define MLX5_FLOW_LAYER_OUTER_L3_IPV6_FRAG_EXT (1u << 30)
 #define MLX5_FLOW_LAYER_INNER_L3_IPV6_FRAG_EXT (1u << 31)
 
+/* Pattern tunnel Layer bits (continued). */
+#define MLX5_FLOW_LAYER_GENEVE_OPT (UINT64_C(1) << 32)
+
 /* Outer Masks. */
 #define MLX5_FLOW_LAYER_OUTER_L3 \
 	(MLX5_FLOW_LAYER_OUTER_L3_IPV4 | MLX5_FLOW_LAYER_OUTER_L3_IPV6)
@@ -1396,6 +1399,11 @@ int mlx5_flow_validate_item_geneve(const struct rte_flow_item *item,
 				   uint64_t item_flags,
 				   struct rte_eth_dev *dev,
 				   struct rte_flow_error *error);
+int mlx5_flow_validate_item_geneve_opt(const struct rte_flow_item *item,
+				   uint64_t last_item,
+				   const struct rte_flow_item *geneve_item,
+				   struct rte_eth_dev *dev,
+				   struct rte_flow_error *error);
 int mlx5_flow_validate_item_ecpri(const struct rte_flow_item *item,
 				  uint64_t item_flags,
 				  uint64_t last_item,
drivers/net/mlx5/mlx5_flow_dv.c +12 -0
index 0a657cea41..ba10b7c706 100644
@@ -5237,6 +5237,7 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr,
 	uint16_t ether_type = 0;
 	int actions_n = 0;
 	uint8_t item_ipv6_proto = 0;
+	const struct rte_flow_item *geneve_item = NULL;
 	const struct rte_flow_item *gre_item = NULL;
 	const struct rte_flow_action_raw_decap *decap;
 	const struct rte_flow_action_raw_encap *encap;
@@ -5518,8 +5519,19 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr,
 							     error);
 			if (ret < 0)
 				return ret;
+			geneve_item = items;
 			last_item = MLX5_FLOW_LAYER_GENEVE;
 			break;
+		case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+			ret = mlx5_flow_validate_item_geneve_opt(items,
+								 last_item,
+								 geneve_item,
+								 dev,
+								 error);
+			if (ret < 0)
+				return ret;
+			last_item = MLX5_FLOW_LAYER_GENEVE_OPT;
+			break;
 		case RTE_FLOW_ITEM_TYPE_MPLS:
 			ret = mlx5_flow_validate_item_mpls(dev, items,
 							   item_flags,
Signature · 1 line
2.21.0
#41 lore raw

PATCH v6 8/9 net/mlx5: add GENEVE TLV option flow translation

In reply to Shiri Kuzin

The GENEVE TLV option matching flows must be created using a translation function.

This function checks whether we already created a Devx object for the matching and either creates the objects or updates the reference counter.

---
 drivers/net/mlx5/mlx5_flow.h    |  2 +-
 drivers/net/mlx5/mlx5_flow_dv.c | 85 +++++++++++++++++++++++++++++++++
 2 files changed, 86 insertions(+), 1 deletion(-)
drivers/net/mlx5/mlx5_flow.h +1 -1
index e52f902448..ed112246ed 100644
@@ -329,7 +329,7 @@ enum mlx5_feature_name {
 #define MLX5_GENEVE_VER_VAL(a) \
 		(((a) >> (MLX5_GENEVE_VER_SHIFT)) & (MLX5_GENEVE_VER_MASK))
 #define MLX5_GENEVE_OPTLEN_MASK 0x3F
-#define MLX5_GENEVE_OPTLEN_SHIFT 7
+#define MLX5_GENEVE_OPTLEN_SHIFT 8
 #define MLX5_GENEVE_OPTLEN_VAL(a) \
 	    (((a) >> (MLX5_GENEVE_OPTLEN_SHIFT)) & (MLX5_GENEVE_OPTLEN_MASK))
 #define MLX5_GENEVE_OAMF_MASK 0x1
drivers/net/mlx5/mlx5_flow_dv.c +85 -0

97 lines not on this page · raw message

Signature · 1 line
2.21.0
#42 lore raw

PATCH v6 9/9 doc: update GENEVE TLV option support

In reply to Shiri Kuzin

GENEVE TLV option support added to mlx5 PMD.

The limitations and support were updated in
documentation.

---
 doc/guides/nics/mlx5.rst               | 23 ++++++++++++++++++++++-
 doc/guides/rel_notes/release_21_02.rst |  8 ++++++++
 2 files changed, 30 insertions(+), 1 deletion(-)
doc/guides/nics/mlx5.rst +22 -1
index 6950cc1188..6d14b890c9 100644
@@ -98,6 +98,7 @@ Features
 - Hardware LRO.
 - Hairpin.
 - Multiple-thread flow insertion.
+- Matching on Geneve TLV option header with raw encap/decap action.
 
 Limitations
 -----------
@@ -175,7 +176,18 @@ Limitations
      - OAM
      - protocol type
      - options length
-       Currently, the only supported options length value is 0.
+
+- Match on Geneve TLv option is supported on the following fields:
+     - Class
+     - Type
+     - Length
+     - Data
+
+  Only one Class/Type/Length Geneve TLV option is supported per shared device.
+  Class/Type/Length fields must be specified as well as masks.
+  Class/Type/Length specified masks must be full.
+  Matching Geneve TLV option without specifying data is not supported.
+  Matching Geneve TLV option with data & mask == 0 is not supported.
 
 - VF: flow rules created on VF devices can only match traffic targeted at the
   configured MAC addresses (see ``rte_eth_dev_mac_addr_add()``).
@@ -1004,6 +1016,10 @@ Below are some firmware configurations listed.
    or
    FLEX_PARSER_PROFILE_ENABLE=1
 
+- enable Geneve TLV option flow matching::
+
+   FLEX_PARSER_PROFILE_ENABLE=0
+
 - enable GTP flow matching::
 
    FLEX_PARSER_PROFILE_ENABLE=3
@@ -1483,6 +1499,11 @@ Supported hardware offloads
    |                       | |  rdma-core 32 | | rdma-core 32  |
    |                       | |  ConnectX-6 Dx| | ConnectX-6 Dx |
    +-----------------------+-----------------+-----------------+
+   | Encapsulation         | | DPDK 21.02    | | DPDK 21.02    |
+   | GENEVE TLV option     | | OFED 5.2      | | OFED 5.2      |
+   |                       | | rdma-core 34  | | rdma-core 34  |
+   |                       | | ConnectX-6 Dx | | ConnectX-6 Dx |
+   +-----------------------+-----------------+-----------------+
 
 Notes for metadata
 ------------------
doc/guides/rel_notes/release_21_02.rst +8 -0
index 706cbf8f0c..1ac0232350 100644
@@ -55,6 +55,14 @@ New Features
      Also, make sure to start the actual text at the margin.
      =======================================================
 
+* **Updated Mellanox mlx5 driver.**
+
+  Updated the Mellanox mlx5 driver with new features and improvements, including:
+
+  * **Added GENEVE TLV option in rte_flow.**
+
+  Added support for matching GENEVE TLV option and raw encap/decap of GENEVE
+  TLV option.
 
 Removed Items
 -------------
Signature · 1 line
2.21.0
#43 lore raw
In reply to Shiri Kuzin
Quoted 48 lines · Shiri Kuzin

The Geneve tunneling protocol is designed to allow the user to specify some data context on the packet.
The GENEVE TLV (Type-Length-Variable) Option is the mean intended to present the user data.

In order to support GENEVE TLV Option the new rte_flow item "rte_flow_item_geneve_opt" is introduced.
The new item contains the values and masks for the following fields:

(quoted diff) +0 -4
-option class
-option type
-length
-data

The usage example:
"flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 / geneve-opt class is 5 length is 1 type is 0 data is 0x66998800 / end actions count / drop / end"

New item will be added to testpmd to support raw encap/decap action.

v6:
- added testpmd support for GENEVE option length.
- updated setting GENEVE option length value.

v5:
- added Devx function to def file.

v4:
- fixed setting option length.
- fixed testpmd calculation of GENEVE option size.
- updated documentation.

v3:
- updated documentation.

v2:
- removed pedantic.

Shiri Kuzin (7):
lib/librte_ethdev: introduce GENEVE header TLV option item common/mlx5: check GENEVE TLV support in HCA attributes common/mlx5: create GENEVE TLV option object with DevX net/mlx5: create GENEVE TLV option management net/mlx5: add GENEVE TLV option flow validation net/mlx5: add GENEVE TLV option flow translation
doc: update GENEVE TLV option support

Quoted 4 lines

Viacheslav Ovsiienko (2):
app/testpmd: add GENEVE option item support app/testpmd: add GENEVE header option length support

Hi Shiri,

Can you please rebase on top of latest next-net, this set conflicts with GTP patchset.

Thanks,
ferruh

#44 lore raw
Quoted 69 lines

-----Original Message-----
From: Ferruh Yigit <[email protected]>
Sent: Friday, January 15, 2021 3:34 AM
To: Shiri Kuzin <[email protected]>; [email protected]
Cc: Slava Ovsiienko <[email protected]>; NBU-Contact-Adrien Mazarguil <[email protected]>; Ori Kam <[email protected]>; NBU-Contact-Thomas Monjalon <[email protected]>; Raslan Darawsheh <[email protected]>; [email protected]
Subject: Re: [PATCH v6 0/9] ethdev: introduce GENEVE header TLV option item

Quoted 48 lines · Shiri Kuzin

The Geneve tunneling protocol is designed to allow the user to specify some data context on the packet.
The GENEVE TLV (Type-Length-Variable) Option is the mean intended to present the user data.

In order to support GENEVE TLV Option the new rte_flow item "rte_flow_item_geneve_opt" is introduced.
The new item contains the values and masks for the following fields:

(quoted diff) +0 -4
-option class
-option type
-length
-data

The usage example:
"flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 / geneve-opt class is 5 length is 1 type is 0 data is 0x66998800 / end actions count / drop / end"

New item will be added to testpmd to support raw encap/decap action.

v6:
- added testpmd support for GENEVE option length.
- updated setting GENEVE option length value.

v5:
- added Devx function to def file.

v4:
- fixed setting option length.
- fixed testpmd calculation of GENEVE option size.
- updated documentation.

v3:
- updated documentation.

v2:
- removed pedantic.

Shiri Kuzin (7):
lib/librte_ethdev: introduce GENEVE header TLV option item common/mlx5: check GENEVE TLV support in HCA attributes common/mlx5: create GENEVE TLV option object with DevX net/mlx5: create GENEVE TLV option management net/mlx5: add GENEVE TLV option flow validation net/mlx5: add GENEVE TLV option flow translation
doc: update GENEVE TLV option support

Quoted 4 lines

Viacheslav Ovsiienko (2):
app/testpmd: add GENEVE option item support app/testpmd: add GENEVE header option length support

Hi Shiri,

Quoted 5 lines

Can you please rebase on top of latest next-net, this set conflicts with GTP patchset.

Thanks,
Ferruh

Hi Ferruh,

Sure I will send updated version.

Thanks,
Shiri.

#45 lore raw

v7 cover ethdev: introduce GENEVE header TLV option item

In reply to Shiri Kuzin

The Geneve tunneling protocol is designed to allow the user to specify some data context on the packet.
The GENEVE TLV (Type-Length-Variable) Option is the mean intended to present the user data.

In order to support GENEVE TLV Option the new rte_flow item "rte_flow_item_geneve_opt" is introduced.
The new item contains the values and masks for the following fields:
-option class
-option type
-length
-data

The usage example:
"flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 / geneve-opt class is 5 length is 1 type is 0 data is 0x66998800 / end actions count / drop / end"

New item will be added to testpmd to support raw encap/decap action.

v7:
- rebased code on top of latest next-net.
- fixed typo in testpmd doc.

v6:
- added testpmd support for GENEVE option length.
- updated setting GENEVE option length value.

v5:
- added Devx function to def file.

v4:
- fixed setting option length.
- fixed testpmd calculation of GENEVE option size.
- updated documentation.

v3:
- updated documentation.

v2:
- removed pedantic.

Shiri Kuzin (7):
lib/librte_ethdev: introduce GENEVE header TLV option item common/mlx5: check GENEVE TLV support in HCA attributes common/mlx5: create GENEVE TLV option object with DevX net/mlx5: create GENEVE TLV option management net/mlx5: add GENEVE TLV option flow validation net/mlx5: add GENEVE TLV option flow translation
doc: update GENEVE TLV option support

Viacheslav Ovsiienko (2):
app/testpmd: add GENEVE option item support app/testpmd: add GENEVE header option length support

 app/test-pmd/cmdline_flow.c                   | 110 +++++++++-
 doc/guides/nics/mlx5.rst                      |  26 ++-
 doc/guides/rel_notes/release_21_02.rst        |   5 +
 doc/guides/testpmd_app_ug/testpmd_funcs.rst   |   8 +
 drivers/common/mlx5/mlx5_devx_cmds.c          |  62 ++++++
 drivers/common/mlx5/mlx5_devx_cmds.h          |   9 +
 drivers/common/mlx5/mlx5_prm.h                |  28 ++-
 .../common/mlx5/rte_common_mlx5_exports.def   |   1 +
 drivers/common/mlx5/version.map               |   1 +
 drivers/net/mlx5/mlx5.c                       |   2 +
 drivers/net/mlx5/mlx5.h                       |  13 ++
 drivers/net/mlx5/mlx5_flow.c                  | 143 ++++++++++++
 drivers/net/mlx5/mlx5_flow.h                  |  13 +-
 drivers/net/mlx5/mlx5_flow_dv.c               | 205 ++++++++++++++++++
 lib/librte_ethdev/rte_flow.c                  |   1 +
 lib/librte_ethdev/rte_flow.h                  |  27 +++
 16 files changed, 643 insertions(+), 11 deletions(-)
Signature · 1 line
2.21.0
#46 lore raw

PATCH v7 1/9 lib/librte_ethdev: introduce GENEVE header TLV option item

The Geneve tunneling protocol is designed to allow the user to specify some data context on the packet. The GENEVE TLV (Type-Length-Variable) Option is the mean intended to present the user data.

In order to support GENEVE TLV Option the new rte_flow item "rte_flow_item_geneve_opt" is added.
The new item contains the values and masks for the following fields:
-option class
-option type
-length
-data

New item will be added to testpmd to support match and raw encap/decap actions.

---
 lib/librte_ethdev/rte_flow.c |  1 +
 lib/librte_ethdev/rte_flow.h | 27 +++++++++++++++++++++++++++
 2 files changed, 28 insertions(+)
lib/librte_ethdev/rte_flow.c +1 -0
index a06f64c271..2af7d965e1 100644
@@ -97,6 +97,7 @@ static const struct rte_flow_desc_data rte_flow_desc_item[] = {
 	MK_FLOW_ITEM(L2TPV3OIP, sizeof(struct rte_flow_item_l2tpv3oip)),
 	MK_FLOW_ITEM(PFCP, sizeof(struct rte_flow_item_pfcp)),
 	MK_FLOW_ITEM(ECPRI, sizeof(struct rte_flow_item_ecpri)),
+	MK_FLOW_ITEM(GENEVE_OPT, sizeof(struct rte_flow_item_geneve_opt)),
 };
 
 /** Generate flow_action[] entry. */
lib/librte_ethdev/rte_flow.h +27 -0
index 0977a78270..11a6494b8e 100644
@@ -543,6 +543,14 @@ enum rte_flow_item_type {
 	 * See struct rte_flow_item_ipv6_frag_ext.
 	 */
 	RTE_FLOW_ITEM_TYPE_IPV6_FRAG_EXT,
+
+	/**
+	 * Matches Geneve Variable Length Option
+	 *
+	 * See struct rte_flow_item_geneve_opt
+	 */
+	RTE_FLOW_ITEM_TYPE_GENEVE_OPT,
+
 };
 
 /**
@@ -1627,6 +1635,25 @@ static const struct rte_flow_item_ecpri rte_flow_item_ecpri_mask = {
 };
 #endif
 
+/**
+ * RTE_FLOW_ITEM_TYPE_GENEVE_OPT
+ *
+ * Matches a GENEVE Variable Length Option
+ */
+struct rte_flow_item_geneve_opt {
+	rte_be16_t option_class;
+	uint8_t option_type;
+	uint8_t option_len;
+	uint32_t *data;
+};
+
+/** Default mask for RTE_FLOW_ITEM_TYPE_GENEVE_OPT. */
+#ifndef __cplusplus
+static const struct rte_flow_item_geneve_opt
+rte_flow_item_geneve_opt_mask = {
+	.option_type = 0xff,
+};
+#endif
 /**
  * Matching pattern item definition.
  *
Signature · 1 line
2.21.0
Reviewed-by on v7 1/9
#47 lore raw

Re: [dpdk-dev] [PATCH v7 1/9] lib/librte_ethdev: introduce GENEVE header TLV option item

Quoted 14 lines · Shiri Kuzin

The Geneve tunneling protocol is designed to allow the user to specify some data context on the packet. The GENEVE TLV (Type-Length-Variable) Option is the mean intended to present the user data.

In order to support GENEVE TLV Option the new rte_flow item "rte_flow_item_geneve_opt" is added.
The new item contains the values and masks for the following fields:

(quoted diff) +0 -4
-option class
-option type
-length
-data
Quoted 5 lines

New item will be added to testpmd to support match and raw encap/decap actions.

#48 lore raw

PATCH v7 2/9 app/testpmd: add GENEVE option item support

In reply to Shiri Kuzin

From: Viacheslav Ovsiienko <[email protected]>

The patch adds the GENEVE option rte flow item support to command line interpreter. The flow command with GENEVE option items looks like:

flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 /
geneve-opt class is 99 length is 1 type is 0 data is 0x669988 / end actions drop / end

The option length should be specified in 32-bit words, this value specifies the length of the data pattern/mask arrays (should be multiplied by sizeof(uint32_t) to be expressed in bytes. If match on the length itself is not needed the mask should be set to zero, in this case length is used to specify the pattern/mask array lengths only.

---
 app/test-pmd/cmdline_flow.c                 | 100 ++++++++++++++++++--
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |   8 ++
 2 files changed, 102 insertions(+), 6 deletions(-)
app/test-pmd/cmdline_flow.c +94 -6
index 37b10e61bf..4e9b76626f 100644
@@ -283,6 +283,11 @@ enum index {
 	ITEM_ECPRI_MSG_IQ_DATA_PCID,
 	ITEM_ECPRI_MSG_RTC_CTRL_RTCID,
 	ITEM_ECPRI_MSG_DLY_MSR_MSRID,
+	ITEM_GENEVE_OPT,
+	ITEM_GENEVE_OPT_CLASS,
+	ITEM_GENEVE_OPT_TYPE,
+	ITEM_GENEVE_OPT_LENGTH,
+	ITEM_GENEVE_OPT_DATA,
 
 	/* Validate/create actions. */
 	ACTIONS,
@@ -413,6 +418,9 @@ enum index {
 /** Maximum size for pattern in struct rte_flow_item_raw. */
 #define ITEM_RAW_PATTERN_SIZE 40
 
+/** Maximum size for GENEVE option data pattern in bytes. */
+#define ITEM_GENEVE_OPT_DATA_SIZE 124
+
 /** Storage size for struct rte_flow_item_raw including pattern. */
 #define ITEM_RAW_SIZE \
 	(sizeof(struct rte_flow_item_raw) + ITEM_RAW_PATTERN_SIZE)
@@ -428,7 +436,7 @@ struct action_rss_data {
 };
 
 /** Maximum data size in struct rte_flow_action_raw_encap. */
-#define ACTION_RAW_ENCAP_MAX_DATA 128
+#define ACTION_RAW_ENCAP_MAX_DATA 512
 #define RAW_ENCAP_CONFS_MAX_NUM 8
 
 /** Storage for struct rte_flow_action_raw_encap. */
@@ -658,6 +666,16 @@ struct token {
 		.mask = (const void *)&(const s){ .f = (1 << (b)) - 1 }, \
 	})
 
+/** Static initializer for ARGS() to target a field with limits. */
+#define ARGS_ENTRY_BOUNDED(s, f, i, a) \
+	(&(const struct arg){ \
+		.bounded = 1, \
+		.min = (i), \
+		.max = (a), \
+		.offset = offsetof(s, f), \
+		.size = sizeof(((s *)0)->f), \
+	})
+
 /** Static initializer for ARGS() to target an arbitrary bit-mask. */
 #define ARGS_ENTRY_MASK(s, f, m) \
 	(&(const struct arg){ \
@@ -903,6 +921,7 @@ static const enum index next_item[] = {
 	ITEM_AH,
 	ITEM_PFCP,
 	ITEM_ECPRI,
+	ITEM_GENEVE_OPT,
 	END_SET,
 	ZERO,
 };
@@ -1244,6 +1263,15 @@ static const enum index item_ecpri_common_type[] = {
 	ZERO,
 };
 
+static const enum index item_geneve_opt[] = {
+	ITEM_GENEVE_OPT_CLASS,
+	ITEM_GENEVE_OPT_TYPE,
+	ITEM_GENEVE_OPT_LENGTH,
+	ITEM_GENEVE_OPT_DATA,
+	ITEM_NEXT,
+	ZERO,
+};
+
 static const enum index next_action[] = {
 	ACTION_END,
 	ACTION_VOID,
@@ -3230,6 +3258,47 @@ static const struct token token_list[] = {
 		.args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_ecpri,
 				hdr.type5.msr_id)),
 	},
+	[ITEM_GENEVE_OPT] = {
+		.name = "geneve-opt",
+		.help = "GENEVE header option",
+		.priv = PRIV_ITEM(GENEVE_OPT,
+				  sizeof(struct rte_flow_item_geneve_opt) +
+				  ITEM_GENEVE_OPT_DATA_SIZE),
+		.next = NEXT(item_geneve_opt),
+		.call = parse_vc,
+	},
+	[ITEM_GENEVE_OPT_CLASS]	= {
+		.name = "class",
+		.help = "GENEVE option class",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(UNSIGNED), item_param),
+		.args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_geneve_opt,
+					     option_class)),
+	},
+	[ITEM_GENEVE_OPT_TYPE] = {
+		.name = "type",
+		.help = "GENEVE option type",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(UNSIGNED), item_param),
+		.args = ARGS(ARGS_ENTRY(struct rte_flow_item_geneve_opt,
+					option_type)),
+	},
+	[ITEM_GENEVE_OPT_LENGTH] = {
+		.name = "length",
+		.help = "GENEVE option data length (in 32b words)",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(UNSIGNED), item_param),
+		.args = ARGS(ARGS_ENTRY_BOUNDED(
+				struct rte_flow_item_geneve_opt, option_len,
+				0, 31)),
+	},
+	[ITEM_GENEVE_OPT_DATA] = {
+		.name = "data",
+		.help = "GENEVE option data pattern",
+		.next = NEXT(item_geneve_opt, NEXT_ENTRY(HEX), item_param),
+		.args = ARGS(ARGS_ENTRY(struct rte_flow_item_geneve_opt, data),
+			     ARGS_ENTRY_ARB(0, 0),
+			     ARGS_ENTRY_ARB
+				(sizeof(struct rte_flow_item_geneve_opt),
+				ITEM_GENEVE_OPT_DATA_SIZE)),
+	},
 	/* Validate/create actions. */
 	[ACTIONS] = {
 		.name = "actions",
@@ -6482,11 +6551,14 @@ parse_hex(struct context *ctx, const struct token *token,
 	ret = snprintf(tmp, sizeof(tmp), "%u", hexlen);
 	if (ret < 0)
 		goto error;
-	push_args(ctx, arg_len);
-	ret = parse_int(ctx, token, tmp, ret, NULL, 0);
-	if (ret < 0) {
-		pop_args(ctx);
-		goto error;
+	/* Save length if requested. */
+	if (arg_len->size) {
+		push_args(ctx, arg_len);
+		ret = parse_int(ctx, token, tmp, ret, NULL, 0);
+		if (ret < 0) {
+			pop_args(ctx);
+			goto error;
+		}
 	}
 	buf = (uint8_t *)ctx->object + arg_data->offset;
 	/* Output buffer is not necessarily NUL-terminated. */
@@ -7486,6 +7558,9 @@ flow_item_default_mask(const struct rte_flow_item *item)
 	case RTE_FLOW_ITEM_TYPE_GENEVE:
 		mask = &rte_flow_item_geneve_mask;
 		break;
+	case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+		mask = &rte_flow_item_geneve_opt_mask;
+		break;
 	case RTE_FLOW_ITEM_TYPE_PPPOE_PROTO_ID:
 		mask = &rte_flow_item_pppoe_proto_id_mask;
 		break;
@@ -7600,6 +7675,7 @@ cmd_set_raw_parsed(const struct buffer *in)
 	data_tail = data + ACTION_RAW_ENCAP_MAX_DATA;
 	for (i = n - 1 ; i >= 0; --i) {
 		const struct rte_flow_item_gtp *gtp;
+		const struct rte_flow_item_geneve_opt *opt;
 
 		item = in->args.vc.pattern + i;
 		if (item->spec == NULL)
@@ -7653,6 +7729,18 @@ cmd_set_raw_parsed(const struct buffer *in)
 		case RTE_FLOW_ITEM_TYPE_GENEVE:
 			size = sizeof(struct rte_geneve_hdr);
 			break;
+		case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+			opt = (const struct rte_flow_item_geneve_opt *)
+								item->spec;
+			size = offsetof(struct rte_flow_item_geneve_opt, data);
+			if (opt->option_len && opt->data) {
+				*total_size += opt->option_len *
+					       sizeof(uint32_t);
+				rte_memcpy(data_tail - (*total_size),
+					   opt->data,
+					   opt->option_len * sizeof(uint32_t));
+			}
+			break;
 		case RTE_FLOW_ITEM_TYPE_L2TPV3OIP:
 			size = sizeof(rte_be32_t);
 			proto = 0x73;
doc/guides/testpmd_app_ug/testpmd_funcs.rst +8 -0
index 842ac42ce6..458f25bb63 100644
@@ -3680,6 +3680,14 @@ This section lists supported pattern items and their attributes, if any.
   - ``vni {unsigned}``: virtual network identifier.
   - ``protocol {unsigned}``: protocol type.
 
+- ``geneve-opt``: match GENEVE header option.
+
+  - ``class {unsigned}``: GENEVE option class.
+  - ``type {unsigned}``: GENEVE option type.
+  - ``length {unsigned}``: GENEVE option length in 32-bit words.
+  - ``data {hex string}``: GENEVE option data, the length is defined by
+    ``length`` field.
+
 - ``vxlan-gpe``: match VXLAN-GPE header.
 
   - ``vni {unsigned}``: VXLAN-GPE identifier.
Signature · 1 line
2.21.0
#49 lore raw

PATCH v7 3/9 app/testpmd: add GENEVE header option length support

In reply to Shiri Kuzin

From: Viacheslav Ovsiienko <[email protected]>

The patch adds the GENEVE rte flow option length support to command line interpreter. The flow command with GENEVE option items looks like:

flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 optlen is 2 / end actions drop / end

The option length should be specified in 32-bit words, this value specifies the all options length in the GENEVE header.

---
 app/test-pmd/cmdline_flow.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
app/test-pmd/cmdline_flow.c +10 -0
index 4e9b76626f..9940d4fff9 100644
@@ -223,6 +223,7 @@ enum index {
 	ITEM_GENEVE,
 	ITEM_GENEVE_VNI,
 	ITEM_GENEVE_PROTO,
+	ITEM_GENEVE_OPTLEN,
 	ITEM_VXLAN_GPE,
 	ITEM_VXLAN_GPE_VNI,
 	ITEM_ARP_ETH_IPV4,
@@ -1101,6 +1102,7 @@ static const enum index item_gtp[] = {
 static const enum index item_geneve[] = {
 	ITEM_GENEVE_VNI,
 	ITEM_GENEVE_PROTO,
+	ITEM_GENEVE_OPTLEN,
 	ITEM_NEXT,
 	ZERO,
 };
@@ -2807,6 +2809,14 @@ static const struct token token_list[] = {
 		.args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_geneve,
 					     protocol)),
 	},
+	[ITEM_GENEVE_OPTLEN] = {
+		.name = "optlen",
+		.help = "GENEVE options length in dwords",
+		.next = NEXT(item_geneve, NEXT_ENTRY(UNSIGNED), item_param),
+		.args = ARGS(ARGS_ENTRY_MASK_HTON(struct rte_flow_item_geneve,
+						  ver_opt_len_o_c_rsvd0,
+						  "\x3f\x00")),
+	},
 	[ITEM_VXLAN_GPE] = {
 		.name = "vxlan-gpe",
 		.help = "match VXLAN-GPE header",
Signature · 1 line
2.21.0
#50 lore raw

PATCH v7 4/9 common/mlx5: check GENEVE TLV support in HCA attributes

In reply to Shiri Kuzin

This is preparation step to support match on GENEVE TLV option.

In this Patch we add the HCA attributes that will allow supporting GENEVE TLV option matching.

---
 drivers/common/mlx5/mlx5_devx_cmds.c |  7 +++++++
 drivers/common/mlx5/mlx5_devx_cmds.h |  4 ++++
 drivers/common/mlx5/mlx5_prm.h       | 28 +++++++++++++++++++++++++---
 3 files changed, 36 insertions(+), 3 deletions(-)
drivers/common/mlx5/mlx5_devx_cmds.c +7 -0
index 4d01f52986..861f75e47d 100644
@@ -690,6 +690,10 @@ mlx5_devx_cmd_query_hca_attr(void *ctx,
 	attr->eth_virt = MLX5_GET(cmd_hca_cap, hcattr, eth_virt);
 	attr->flex_parser_protocols = MLX5_GET(cmd_hca_cap, hcattr,
 					       flex_parser_protocols);
+	attr->max_geneve_tlv_options = MLX5_GET(cmd_hca_cap, hcattr,
+			max_geneve_tlv_options);
+	attr->max_geneve_tlv_option_data_len = MLX5_GET(cmd_hca_cap, hcattr,
+			max_geneve_tlv_option_data_len);
 	attr->qos.sup = MLX5_GET(cmd_hca_cap, hcattr, qos);
 	attr->vdpa.valid = !!(MLX5_GET64(cmd_hca_cap, hcattr,
 					 general_obj_types) &
@@ -717,6 +721,9 @@ mlx5_devx_cmd_query_hca_attr(void *ctx,
 	attr->flow_hit_aso = !!(MLX5_GET64(cmd_hca_cap, hcattr,
 					   general_obj_types) &
 				MLX5_GENERAL_OBJ_TYPES_CAP_FLOW_HIT_ASO);
+	attr->geneve_tlv_opt = !!(MLX5_GET64(cmd_hca_cap, hcattr,
+					   general_obj_types) &
+				MLX5_GENERAL_OBJ_TYPES_CAP_GENEVE_TLV_OPT);
 	attr->log_max_cq = MLX5_GET(cmd_hca_cap, hcattr, log_max_cq);
 	attr->log_max_qp = MLX5_GET(cmd_hca_cap, hcattr, log_max_qp);
 	attr->log_max_cq_sz = MLX5_GET(cmd_hca_cap, hcattr, log_max_cq_sz);
drivers/common/mlx5/mlx5_devx_cmds.h +4 -0
index 8d993dfad7..1e0a48d810 100644
@@ -97,6 +97,8 @@ struct mlx5_hca_attr {
 	uint32_t lro_timer_supported_periods[MLX5_LRO_NUM_SUPP_PERIODS];
 	uint16_t lro_min_mss_size;
 	uint32_t flex_parser_protocols;
+	uint32_t max_geneve_tlv_options;
+	uint32_t max_geneve_tlv_option_data_len;
 	uint32_t hairpin:1;
 	uint32_t log_max_hairpin_queues:5;
 	uint32_t log_max_hairpin_wq_data_sz:5;
@@ -116,6 +118,7 @@ struct mlx5_hca_attr {
 	uint32_t regex:1;
 	uint32_t regexp_num_of_engines;
 	uint32_t log_max_ft_sampler_num:8;
+	uint32_t geneve_tlv_opt;
 	struct mlx5_hca_qos_attr qos;
 	struct mlx5_hca_vdpa_attr vdpa;
 	int log_max_qp_sz;
@@ -481,6 +484,7 @@ struct mlx5_devx_obj *mlx5_devx_cmd_create_flex_parser(void *ctx,
 __rte_internal
 int mlx5_devx_cmd_register_read(void *ctx, uint16_t reg_id,
 				uint32_t arg, uint32_t *data, uint32_t dw_cnt);
+
 /**
  * Create virtio queue counters object DevX API.
  *
drivers/common/mlx5/mlx5_prm.h +25 -3
index ad07e79d78..c9eba22ad7 100644
@@ -789,7 +789,7 @@ struct mlx5_ifc_fte_match_set_misc3_bits {
 	u8 icmp_code[0x8];
 	u8 icmpv6_type[0x8];
 	u8 icmpv6_code[0x8];
-	u8 reserved_at_120[0x20];
+	u8 geneve_tlv_option_0_data[0x20];
 	u8 gtpu_teid[0x20];
 	u8 gtpu_msg_type[0x08];
 	u8 gtpu_msg_flags[0x08];
@@ -1087,6 +1087,8 @@ enum {
 			(1ULL << MLX5_GENERAL_OBJ_TYPE_FLEX_PARSE_GRAPH)
 #define MLX5_GENERAL_OBJ_TYPES_CAP_FLOW_HIT_ASO \
 			(1ULL << MLX5_GENERAL_OBJ_TYPE_FLOW_HIT_ASO)
+#define MLX5_GENERAL_OBJ_TYPES_CAP_GENEVE_TLV_OPT \
+			(1ULL << MLX5_OBJ_TYPE_GENEVE_TLV_OPT)
 
 enum {
 	MLX5_HCA_CAP_OPMOD_GET_MAX   = 0,
@@ -1385,8 +1387,10 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 	u8 reserved_at_500[0x20];
 	u8 num_of_uars_per_page[0x20];
 	u8 flex_parser_protocols[0x20];
-	u8 reserved_at_560[0x20];
-	u8 reserved_at_580[0x3c];
+	u8 max_geneve_tlv_options[0x8];
+	u8 reserved_at_568[0x3];
+	u8 max_geneve_tlv_option_data_len[0x5];
+	u8 reserved_at_570[0x4c];
 	u8 mini_cqe_resp_stride_index[0x1];
 	u8 cqe_128_always[0x1];
 	u8 cqe_compression_128[0x1];
@@ -2297,6 +2301,7 @@ struct mlx5_ifc_create_cq_in_bits {
 };
 
 enum {
+	MLX5_OBJ_TYPE_GENEVE_TLV_OPT = 0x000b,
 	MLX5_GENERAL_OBJ_TYPE_VIRTQ = 0x000d,
 	MLX5_GENERAL_OBJ_TYPE_VIRTIO_Q_COUNTERS = 0x001c,
 	MLX5_GENERAL_OBJ_TYPE_FLEX_PARSE_GRAPH = 0x0022,
@@ -2331,6 +2336,17 @@ struct mlx5_ifc_virtio_q_counters_bits {
 	u8 reserved_at_180[0x50];
 };
 
+struct mlx5_ifc_geneve_tlv_option_bits {
+	u8 modify_field_select[0x40];
+	u8 reserved_at_40[0x18];
+	u8 geneve_option_fte_index[0x8];
+	u8 option_class[0x10];
+	u8 option_type[0x8];
+	u8 reserved_at_78[0x3];
+	u8 option_data_length[0x5];
+	u8 reserved_at_80[0x180];
+};
+
 struct mlx5_ifc_create_virtio_q_counters_in_bits {
 	struct mlx5_ifc_general_obj_in_cmd_hdr_bits hdr;
 	struct mlx5_ifc_virtio_q_counters_bits virtio_q_counters;
@@ -2340,6 +2356,12 @@ struct mlx5_ifc_query_virtio_q_counters_out_bits {
 	struct mlx5_ifc_general_obj_in_cmd_hdr_bits hdr;
 	struct mlx5_ifc_virtio_q_counters_bits virtio_q_counters;
 };
+
+struct mlx5_ifc_create_geneve_tlv_option_in_bits {
+	struct mlx5_ifc_general_obj_in_cmd_hdr_bits hdr;
+	struct mlx5_ifc_geneve_tlv_option_bits geneve_tlv_opt;
+};
+
 enum {
 	MLX5_VIRTQ_STATE_INIT = 0,
 	MLX5_VIRTQ_STATE_RDY = 1,
Signature · 1 line
2.21.0
#51 lore raw

PATCH v7 5/9 common/mlx5: create GENEVE TLV option object with DevX

In reply to Shiri Kuzin

TLV object is a special firmware maintained entity used to support match on GENEVE header extension option.

The TLV object is created with DevX API and accepts the option class, type and lehgth fields.

The class type and length fields are set using MLX5_SET and the Devx object is created using mlx5 glue function.

---
 drivers/common/mlx5/mlx5_devx_cmds.c          | 55 +++++++++++++++++++
 drivers/common/mlx5/mlx5_devx_cmds.h          |  5 ++
 .../common/mlx5/rte_common_mlx5_exports.def   |  1 +
 drivers/common/mlx5/version.map               |  1 +
 4 files changed, 62 insertions(+)
drivers/common/mlx5/mlx5_devx_cmds.c +55 -0
index 861f75e47d..d5859c2a26 100644
@@ -2093,3 +2093,58 @@ mlx5_devx_cmd_alloc_pd(void *ctx)
 	ppd->id = MLX5_GET(alloc_pd_out, out, pd);
 	return ppd;
 }
+
+/**
+ * Create general object of type GENEVE TLV option using DevX API.
+ *
+ * @param[in] ctx
+ *   Context returned from mlx5 open_device() glue function.
+ * @param [in] class
+ *   TLV option variable value of class
+ * @param [in] type
+ *   TLV option variable value of type
+ * @param [in] len
+ *   TLV option variable value of len
+ *
+ * @return
+ *   The DevX object created, NULL otherwise and rte_errno is set.
+ */
+struct mlx5_devx_obj *
+mlx5_devx_cmd_create_geneve_tlv_option(void *ctx,
+		uint16_t class, uint8_t type, uint8_t len)
+{
+	uint32_t in[MLX5_ST_SZ_DW(create_geneve_tlv_option_in)] = {0};
+	uint32_t out[MLX5_ST_SZ_DW(general_obj_out_cmd_hdr)] = {0};
+	struct mlx5_devx_obj *geneve_tlv_opt_obj = mlx5_malloc(MLX5_MEM_ZERO,
+						   sizeof(*geneve_tlv_opt_obj),
+						   0, SOCKET_ID_ANY);
+
+	if (!geneve_tlv_opt_obj) {
+		DRV_LOG(ERR, "Failed to allocate geneve tlv option object.");
+		rte_errno = ENOMEM;
+		return NULL;
+	}
+	void *hdr = MLX5_ADDR_OF(create_geneve_tlv_option_in, in, hdr);
+	void *opt = MLX5_ADDR_OF(create_geneve_tlv_option_in, in,
+			geneve_tlv_opt);
+	MLX5_SET(general_obj_in_cmd_hdr, hdr, opcode,
+			MLX5_CMD_OP_CREATE_GENERAL_OBJECT);
+	MLX5_SET(general_obj_in_cmd_hdr, hdr, obj_type,
+			MLX5_OBJ_TYPE_GENEVE_TLV_OPT);
+	MLX5_SET(geneve_tlv_option, opt, option_class,
+			rte_be_to_cpu_16(class));
+	MLX5_SET(geneve_tlv_option, opt, option_type, type);
+	MLX5_SET(geneve_tlv_option, opt, option_data_length, len);
+	geneve_tlv_opt_obj->obj = mlx5_glue->devx_obj_create(ctx, in,
+					sizeof(in), out, sizeof(out));
+	if (!geneve_tlv_opt_obj->obj) {
+		rte_errno = errno;
+		DRV_LOG(ERR, "Failed to create Geneve tlv option "
+				"Obj using DevX.");
+		mlx5_free(geneve_tlv_opt_obj);
+		return NULL;
+	}
+	geneve_tlv_opt_obj->id = MLX5_GET(general_obj_out_cmd_hdr, out, obj_id);
+	return geneve_tlv_opt_obj;
+}
+
drivers/common/mlx5/mlx5_devx_cmds.h +5 -0
index 1e0a48d810..bf83a903b1 100644
@@ -485,6 +485,11 @@ __rte_internal
 int mlx5_devx_cmd_register_read(void *ctx, uint16_t reg_id,
 				uint32_t arg, uint32_t *data, uint32_t dw_cnt);
 
+__rte_internal
+struct mlx5_devx_obj *
+mlx5_devx_cmd_create_geneve_tlv_option(void *ctx,
+		uint16_t class, uint8_t type, uint8_t len);
+
 /**
  * Create virtio queue counters object DevX API.
  *
drivers/common/mlx5/rte_common_mlx5_exports.def +1 -0
index b385d38d1a..fd62b806ca 100644
@@ -34,6 +34,7 @@ EXPORTS
 	mlx5_devx_cmd_register_read
 	mlx5_devx_get_out_command_status
 	mlx5_devx_cmd_create_flow_hit_aso_obj
+	mlx5_devx_cmd_create_geneve_tlv_option
 
 	mlx5_devx_cq_create
 	mlx5_devx_cq_destroy
drivers/common/mlx5/version.map +1 -0
index 00760abece..244b9c7339 100644
@@ -23,6 +23,7 @@ INTERNAL {
 	mlx5_devx_cmd_create_virtio_q_counters;
 	mlx5_devx_cmd_create_virtq;
         mlx5_devx_cmd_create_flow_hit_aso_obj;
+	mlx5_devx_cmd_create_geneve_tlv_option;
 	mlx5_devx_cmd_destroy;
 	mlx5_devx_cmd_flow_counter_alloc;
 	mlx5_devx_cmd_flow_counter_query;
Signature · 1 line
2.21.0
#52 lore raw

PATCH v7 6/9 net/mlx5: create GENEVE TLV option management

In reply to Shiri Kuzin

Currently firmware supports the only TLV object per device to match on the GENEVE header option.

This patch adds the simple TLV object management to the mlx5 PMD.

---
 drivers/net/mlx5/mlx5.c         |   2 +
 drivers/net/mlx5/mlx5.h         |  13 ++++
 drivers/net/mlx5/mlx5_flow.h    |   5 ++
 drivers/net/mlx5/mlx5_flow_dv.c | 108 ++++++++++++++++++++++++++++++++
 4 files changed, 128 insertions(+)
drivers/net/mlx5/mlx5.c +2 -0
index e245276fce..3730f32295 100644
@@ -1008,6 +1008,7 @@ mlx5_alloc_shared_dev_ctx(const struct mlx5_dev_spawn_data *spawn,
 	rte_rwlock_write_unlock(&mlx5_shared_data->mem_event_rwlock);
 	/* Add context to the global device list. */
 	LIST_INSERT_HEAD(&mlx5_dev_ctx_list, sh, next);
+	rte_spinlock_init(&sh->geneve_tlv_opt_sl);
 exit:
 	pthread_mutex_unlock(&mlx5_dev_ctx_list_mutex);
 	return sh;
@@ -1106,6 +1107,7 @@ mlx5_free_shared_dev_ctx(struct mlx5_dev_ctx_shared *sh)
 		mlx5_glue->devx_free_uar(sh->devx_rx_uar);
 	if (sh->ctx)
 		claim_zero(mlx5_glue->close_device(sh->ctx));
+	MLX5_ASSERT(sh->geneve_tlv_option_resource == NULL);
 	pthread_mutex_destroy(&sh->txpp.mutex);
 	mlx5_free(sh);
 	return;
drivers/net/mlx5/mlx5.h +13 -0
index 3836a9696c..101e9c20d0 100644
@@ -528,6 +528,16 @@ struct mlx5_aso_age_mng {
 	struct mlx5_aso_sq aso_sq; /* ASO queue objects. */
 };
 
+/* Management structure for geneve tlv option */
+struct mlx5_geneve_tlv_option_resource {
+	struct mlx5_devx_obj *obj; /* Pointer to the geneve tlv opt object. */
+	rte_be16_t option_class; /* geneve tlv opt class.*/
+	uint8_t option_type; /* geneve tlv opt type.*/
+	uint8_t length; /* geneve tlv opt length. */
+	uint32_t refcnt; /* geneve tlv object reference counter */
+};
+
+
 #define MLX5_AGE_EVENT_NEW		1
 #define MLX5_AGE_TRIGGER		2
 #define MLX5_AGE_SET(age_info, BIT) \
@@ -727,6 +737,9 @@ struct mlx5_dev_ctx_shared {
 	void *devx_rx_uar; /* DevX UAR for Rx. */
 	struct mlx5_aso_age_mng *aso_age_mng;
 	/* Management data for aging mechanism using ASO Flow Hit. */
+	struct mlx5_geneve_tlv_option_resource *geneve_tlv_option_resource;
+	/* Management structure for geneve tlv option */
+	rte_spinlock_t geneve_tlv_opt_sl; /* Lock for geneve tlv resource */
 	struct mlx5_dev_shared_port port[]; /* per device port data array. */
 };
 
drivers/net/mlx5/mlx5_flow.h +5 -0
index eb9f97a278..0c708b74e8 100644
@@ -1057,6 +1057,7 @@ struct rte_flow {
 	uint32_t counter; /**< Holds flow counter. */
 	uint32_t tunnel_id;  /**< Tunnel id */
 	uint32_t age; /**< Holds ASO age bit index. */
+	uint32_t geneve_tlv_option; /**< Holds Geneve TLV option id. > */
 } __rte_packed;
 
 /*
@@ -1514,6 +1515,9 @@ void flow_dv_dest_array_remove_cb(struct mlx5_cache_list *list,
 				  struct mlx5_cache_entry *entry);
 struct mlx5_aso_age_action *flow_aso_age_get_by_idx(struct rte_eth_dev *dev,
 						    uint32_t age_idx);
+int flow_dev_geneve_tlv_option_resource_register(struct rte_eth_dev *dev,
+					     const struct rte_flow_item *item,
+					     struct rte_flow_error *error);
 
 void flow_release_workspace(void *data);
 int mlx5_flow_os_init_workspace_once(void);
@@ -1521,4 +1525,5 @@ void *mlx5_flow_os_get_specific_workspace(void);
 int mlx5_flow_os_set_specific_workspace(struct mlx5_flow_workspace *data);
 void mlx5_flow_os_release_workspace(void);
 
+
 #endif /* RTE_PMD_MLX5_FLOW_H_ */
drivers/net/mlx5/mlx5_flow_dv.c +108 -0

126 lines not on this page · raw message

Signature · 1 line
2.21.0
#53 lore raw

PATCH v7 7/9 net/mlx5: add GENEVE TLV option flow validation

In reply to Shiri Kuzin

This patch adds validation routine for the GENEVE header TLV option.

The GENEVE TLV option match must include all fields with full masks due to NIC does not support masking on option class, type and length.

The option data length must be non zero and provided data pattern should be zero neither due to hardware limitations.

---
 drivers/net/mlx5/mlx5_flow.c    | 143 ++++++++++++++++++++++++++++++++
 drivers/net/mlx5/mlx5_flow.h    |   6 ++
 drivers/net/mlx5/mlx5_flow_dv.c |  12 +++
 3 files changed, 161 insertions(+)
drivers/net/mlx5/mlx5_flow.c +143 -0
index 1790e774ba..3c4ef6d7d2 100644
@@ -2622,6 +2622,149 @@ mlx5_flow_validate_item_geneve(const struct rte_flow_item *item,
 	return 0;
 }
 
+/**
+ * Validate Geneve TLV option item.
+ *
+ * @param[in] item
+ *   Item specification.
+ * @param[in] last_item
+ *   Previous validated item in the pattern items.
+ * @param[in] geneve_item
+ *   Previous GENEVE item specification.
+ * @param[in] dev
+ *   Pointer to the rte_eth_dev structure.
+ * @param[out] error
+ *   Pointer to error structure.
+ *
+ * @return
+ *   0 on success, a negative errno value otherwise and rte_errno is set.
+ */
+int
+mlx5_flow_validate_item_geneve_opt(const struct rte_flow_item *item,
+				   uint64_t last_item,
+				   const struct rte_flow_item *geneve_item,
+				   struct rte_eth_dev *dev,
+				   struct rte_flow_error *error)
+{
+	struct mlx5_priv *priv = dev->data->dev_private;
+	struct mlx5_dev_ctx_shared *sh = priv->sh;
+	struct mlx5_geneve_tlv_option_resource *geneve_opt_resource;
+	struct mlx5_hca_attr *hca_attr = &priv->config.hca_attr;
+	uint8_t data_max_supported =
+			hca_attr->max_geneve_tlv_option_data_len * 4;
+	struct mlx5_dev_config *config = &priv->config;
+	const struct rte_flow_item_geneve *geneve_spec;
+	const struct rte_flow_item_geneve *geneve_mask;
+	const struct rte_flow_item_geneve_opt *spec = item->spec;
+	const struct rte_flow_item_geneve_opt *mask = item->mask;
+	unsigned int i;
+	unsigned int data_len;
+	uint8_t tlv_option_len;
+	uint16_t optlen_m, optlen_v;
+	const struct rte_flow_item_geneve_opt full_mask = {
+		.option_class = RTE_BE16(0xffff),
+		.option_type = 0xff,
+		.option_len = 0x1f,
+	};
+
+	if (!mask)
+		mask = &rte_flow_item_geneve_opt_mask;
+	if (!spec)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt class/type/length must be specified");
+	if ((uint32_t)spec->option_len > MLX5_GENEVE_OPTLEN_MASK)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt length exceeeds the limit (31)");
+	/* Check if class type and length masks are full. */
+	if (full_mask.option_class != mask->option_class ||
+	    full_mask.option_type != mask->option_type ||
+	    full_mask.option_len != (mask->option_len & full_mask.option_len))
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt class/type/length masks must be full");
+	/* Check if length is supported */
+	if ((uint32_t)spec->option_len >
+			config->hca_attr.max_geneve_tlv_option_data_len)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt length not supported");
+	if (config->hca_attr.max_geneve_tlv_options > 1)
+		DRV_LOG(DEBUG,
+			"max_geneve_tlv_options supports more than 1 option");
+	/* Check GENEVE item preceding. */
+	if (!geneve_item || !(last_item & MLX5_FLOW_LAYER_GENEVE))
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve opt item must be preceded with Geneve item");
+	geneve_spec = geneve_item->spec;
+	geneve_mask = geneve_item->mask ? geneve_item->mask :
+					  &rte_flow_item_geneve_mask;
+	/* Check if GENEVE TLV option size doesn't exceed option length */
+	if (geneve_spec && (geneve_mask->ver_opt_len_o_c_rsvd0 ||
+			    geneve_spec->ver_opt_len_o_c_rsvd0)) {
+		tlv_option_len = spec->option_len & mask->option_len;
+		optlen_v = rte_be_to_cpu_16(geneve_spec->ver_opt_len_o_c_rsvd0);
+		optlen_v = MLX5_GENEVE_OPTLEN_VAL(optlen_v);
+		optlen_m = rte_be_to_cpu_16(geneve_mask->ver_opt_len_o_c_rsvd0);
+		optlen_m = MLX5_GENEVE_OPTLEN_VAL(optlen_m);
+		if ((optlen_v & optlen_m) <= tlv_option_len)
+			return rte_flow_error_set
+				(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+				 "GENEVE TLV option length exceeds optlen");
+	}
+	/* Check if length is 0 or data is 0. */
+	if (spec->data == NULL || spec->option_len == 0)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt with zero data/length not supported");
+	/* Check not all data & mask are 0. */
+	data_len = spec->option_len * 4;
+	if (mask->data == NULL) {
+		for (i = 0; i < data_len; i++)
+			if (spec->data[i])
+				break;
+		if (i == data_len)
+			return rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_ITEM, item,
+				"Can't match on Geneve option data 0");
+	} else {
+		for (i = 0; i < data_len; i++)
+			if (spec->data[i] & mask->data[i])
+				break;
+		if (i == data_len)
+			return rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_ITEM, item,
+				"Can't match on Geneve option data and mask 0");
+		/* Check data mask supported. */
+		for (i = data_max_supported; i < data_len ; i++)
+			if (mask->data[i])
+				return rte_flow_error_set(error, ENOTSUP,
+					RTE_FLOW_ERROR_TYPE_ITEM, item,
+					"Data mask is of unsupported size");
+	}
+	/* Check GENEVE option is supported in NIC. */
+	if (!config->hca_attr.geneve_tlv_opt)
+		return rte_flow_error_set
+			(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item,
+			"Geneve TLV opt not supported");
+	/* Check if we already have geneve option with different type/class. */
+	rte_spinlock_lock(&sh->geneve_tlv_opt_sl);
+	geneve_opt_resource = sh->geneve_tlv_option_resource;
+	if (geneve_opt_resource != NULL)
+		if (geneve_opt_resource->option_class != spec->option_class ||
+		    geneve_opt_resource->option_type != spec->option_type ||
+		    geneve_opt_resource->length != spec->option_len) {
+			rte_spinlock_unlock(&sh->geneve_tlv_opt_sl);
+			return rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_ITEM, item,
+				"Only one Geneve TLV option supported");
+		}
+	rte_spinlock_unlock(&sh->geneve_tlv_opt_sl);
+	return 0;
+}
+
 /**
  * Validate MPLS item.
  *
drivers/net/mlx5/mlx5_flow.h +6 -0
index 0c708b74e8..4c68c65f4f 100644
@@ -139,6 +139,7 @@ enum mlx5_feature_name {
 #define MLX5_FLOW_LAYER_INNER_L3_IPV6_FRAG_EXT (1u << 31)
 
 /* Pattern tunnel Layer bits (continued). */
+#define MLX5_FLOW_LAYER_GENEVE_OPT (UINT64_C(1) << 32)
 #define MLX5_FLOW_LAYER_GTP_PSC (UINT64_C(1) << 33)
 
 /* Outer Masks. */
@@ -1407,6 +1408,11 @@ int mlx5_flow_validate_item_geneve(const struct rte_flow_item *item,
 				   uint64_t item_flags,
 				   struct rte_eth_dev *dev,
 				   struct rte_flow_error *error);
+int mlx5_flow_validate_item_geneve_opt(const struct rte_flow_item *item,
+				   uint64_t last_item,
+				   const struct rte_flow_item *geneve_item,
+				   struct rte_eth_dev *dev,
+				   struct rte_flow_error *error);
 int mlx5_flow_validate_item_ecpri(const struct rte_flow_item *item,
 				  uint64_t item_flags,
 				  uint64_t last_item,
drivers/net/mlx5/mlx5_flow_dv.c +12 -0
index 2c5ccd06b7..86c9d2cbbd 100644
@@ -5303,6 +5303,7 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr,
 	uint16_t ether_type = 0;
 	int actions_n = 0;
 	uint8_t item_ipv6_proto = 0;
+	const struct rte_flow_item *geneve_item = NULL;
 	const struct rte_flow_item *gre_item = NULL;
 	const struct rte_flow_item *gtp_item = NULL;
 	const struct rte_flow_action_raw_decap *decap;
@@ -5585,8 +5586,19 @@ flow_dv_validate(struct rte_eth_dev *dev, const struct rte_flow_attr *attr,
 							     error);
 			if (ret < 0)
 				return ret;
+			geneve_item = items;
 			last_item = MLX5_FLOW_LAYER_GENEVE;
 			break;
+		case RTE_FLOW_ITEM_TYPE_GENEVE_OPT:
+			ret = mlx5_flow_validate_item_geneve_opt(items,
+								 last_item,
+								 geneve_item,
+								 dev,
+								 error);
+			if (ret < 0)
+				return ret;
+			last_item = MLX5_FLOW_LAYER_GENEVE_OPT;
+			break;
 		case RTE_FLOW_ITEM_TYPE_MPLS:
 			ret = mlx5_flow_validate_item_mpls(dev, items,
 							   item_flags,
Signature · 1 line
2.21.0
#54 lore raw

PATCH v7 8/9 net/mlx5: add GENEVE TLV option flow translation

In reply to Shiri Kuzin

The GENEVE TLV option matching flows must be created using a translation function.

This function checks whether we already created a Devx object for the matching and either creates the objects or updates the reference counter.

---
 drivers/net/mlx5/mlx5_flow.h    |  2 +-
 drivers/net/mlx5/mlx5_flow_dv.c | 85 +++++++++++++++++++++++++++++++++
 2 files changed, 86 insertions(+), 1 deletion(-)
drivers/net/mlx5/mlx5_flow.h +1 -1
index 4c68c65f4f..a9e33b8ab8 100644
@@ -330,7 +330,7 @@ enum mlx5_feature_name {
 #define MLX5_GENEVE_VER_VAL(a) \
 		(((a) >> (MLX5_GENEVE_VER_SHIFT)) & (MLX5_GENEVE_VER_MASK))
 #define MLX5_GENEVE_OPTLEN_MASK 0x3F
-#define MLX5_GENEVE_OPTLEN_SHIFT 7
+#define MLX5_GENEVE_OPTLEN_SHIFT 8
 #define MLX5_GENEVE_OPTLEN_VAL(a) \
 	    (((a) >> (MLX5_GENEVE_OPTLEN_SHIFT)) & (MLX5_GENEVE_OPTLEN_MASK))
 #define MLX5_GENEVE_OAMF_MASK 0x1
drivers/net/mlx5/mlx5_flow_dv.c +85 -0

97 lines not on this page · raw message

Signature · 1 line
2.21.0
#55 lore raw

PATCH v7 9/9 doc: update GENEVE TLV option support

In reply to Shiri Kuzin

GENEVE TLV option support added to mlx5 PMD.

The limitations and support were updated in
documentation.

---
 doc/guides/nics/mlx5.rst               | 26 +++++++++++++++++++++++++-
 doc/guides/rel_notes/release_21_02.rst |  5 +++++
 2 files changed, 30 insertions(+), 1 deletion(-)
doc/guides/nics/mlx5.rst +25 -1
index db0c8b6c20..17f9f49e83 100644
@@ -98,7 +98,11 @@ Features
 - Hardware LRO.
 - Hairpin.
 - Multiple-thread flow insertion.
+<<<<<<< HEAD
 - Matching on GTP extension header with raw encap/decap action.
+=======
+- Matching on Geneve TLV option header with raw encap/decap action.
+>>>>>>> a2594fbf3c... doc: update GENEVE TLV option support
 
 Limitations
 -----------
@@ -186,7 +190,18 @@ Limitations
      - OAM
      - protocol type
      - options length
-       Currently, the only supported options length value is 0.
+
+- Match on Geneve TLv option is supported on the following fields:
+     - Class
+     - Type
+     - Length
+     - Data
+
+  Only one Class/Type/Length Geneve TLV option is supported per shared device.
+  Class/Type/Length fields must be specified as well as masks.
+  Class/Type/Length specified masks must be full.
+  Matching Geneve TLV option without specifying data is not supported.
+  Matching Geneve TLV option with data & mask == 0 is not supported.
 
 - VF: flow rules created on VF devices can only match traffic targeted at the
   configured MAC addresses (see ``rte_eth_dev_mac_addr_add()``).
@@ -1019,6 +1034,10 @@ Below are some firmware configurations listed.
    or
    FLEX_PARSER_PROFILE_ENABLE=1
 
+- enable Geneve TLV option flow matching::
+
+   FLEX_PARSER_PROFILE_ENABLE=0
+
 - enable GTP flow matching::
 
    FLEX_PARSER_PROFILE_ENABLE=3
@@ -1540,6 +1559,11 @@ Supported hardware offloads
    |                       | |  rdma-core 35 | | rdma-core 35  |
    |                       | |  ConnectX-6 Dx| | ConnectX-6 Dx |
    +-----------------------+-----------------+-----------------+
+   | Encapsulation         | | DPDK 21.02    | | DPDK 21.02    |
+   | GENEVE TLV option     | | OFED 5.2      | | OFED 5.2      |
+   |                       | | rdma-core 34  | | rdma-core 34  |
+   |                       | | ConnectX-6 Dx | | ConnectX-6 Dx |
+   +-----------------------+-----------------+-----------------+
 
 Notes for metadata
 ------------------
doc/guides/rel_notes/release_21_02.rst +5 -0
index 3bb0b9a9ed..066547b6ad 100644
@@ -68,6 +68,11 @@ New Features
   * Introduced basic support on Windows.
   * Added GTP PDU session container matching and raw encap/decap.
 
+* **Added GENEVE TLV option in rte_flow.**
+
+  Added support for matching GENEVE TLV option and raw encap/decap of GENEVE
+  TLV option.
+
 * **Updated GSO support.**
 
   * Added inner UDP/IPv4 support for VXLAN IPv4 GSO.
Signature · 1 line
2.21.0
#56 lore raw
In reply to Shiri Kuzin
Quoted 52 lines · Shiri Kuzin

The Geneve tunneling protocol is designed to allow the user to specify some data context on the packet.
The GENEVE TLV (Type-Length-Variable) Option is the mean intended to present the user data.

In order to support GENEVE TLV Option the new rte_flow item "rte_flow_item_geneve_opt" is introduced.
The new item contains the values and masks for the following fields:

(quoted diff) +0 -4
-option class
-option type
-length
-data

The usage example:
"flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 / geneve-opt class is 5 length is 1 type is 0 data is 0x66998800 / end actions count / drop / end"

New item will be added to testpmd to support raw encap/decap action.

v7:
- rebased code on top of latest next-net.
- fixed typo in testpmd doc.

v6:
- added testpmd support for GENEVE option length.
- updated setting GENEVE option length value.

v5:
- added Devx function to def file.

v4:
- fixed setting option length.
- fixed testpmd calculation of GENEVE option size.
- updated documentation.

v3:
- updated documentation.

v2:
- removed pedantic.

Shiri Kuzin (7):
lib/librte_ethdev: introduce GENEVE header TLV option item common/mlx5: check GENEVE TLV support in HCA attributes common/mlx5: create GENEVE TLV option object with DevX net/mlx5: create GENEVE TLV option management net/mlx5: add GENEVE TLV option flow validation net/mlx5: add GENEVE TLV option flow translation
doc: update GENEVE TLV option support

Quoted 4 lines

Viacheslav Ovsiienko (2):
app/testpmd: add GENEVE option item support app/testpmd: add GENEVE header option length support

Series applied to dpdk-next-net/main, thanks.

#57 lore raw
In reply to Shiri Kuzin
Quoted 52 lines · Shiri Kuzin

The Geneve tunneling protocol is designed to allow the user to specify some data context on the packet.
The GENEVE TLV (Type-Length-Variable) Option is the mean intended to present the user data.

In order to support GENEVE TLV Option the new rte_flow item "rte_flow_item_geneve_opt" is introduced.
The new item contains the values and masks for the following fields:

(quoted diff) +0 -4
-option class
-option type
-length
-data

The usage example:
"flow create 0 ingress pattern eth / ipv4 / udp / geneve vni is 100 / geneve-opt class is 5 length is 1 type is 0 data is 0x66998800 / end actions count / drop / end"

New item will be added to testpmd to support raw encap/decap action.

v7:
- rebased code on top of latest next-net.
- fixed typo in testpmd doc.

v6:
- added testpmd support for GENEVE option length.
- updated setting GENEVE option length value.

v5:
- added Devx function to def file.

v4:
- fixed setting option length.
- fixed testpmd calculation of GENEVE option size.
- updated documentation.

v3:
- updated documentation.

v2:
- removed pedantic.

Shiri Kuzin (7):
lib/librte_ethdev: introduce GENEVE header TLV option item common/mlx5: check GENEVE TLV support in HCA attributes common/mlx5: create GENEVE TLV option object with DevX net/mlx5: create GENEVE TLV option management net/mlx5: add GENEVE TLV option flow validation net/mlx5: add GENEVE TLV option flow translation
doc: update GENEVE TLV option support

Quoted 4 lines

Viacheslav Ovsiienko (2):
app/testpmd: add GENEVE option item support app/testpmd: add GENEVE header option length support

Series applied to dpdk-next-net/main, thanks.