From: <kirankumark@marvell.com>
To: Jerin Jacob <jerinj@marvell.com>,
Nithin Dabilpuram <ndabilpuram@marvell.com>,
Kiran Kumar K <kirankumark@marvell.com>,
"John McNamara" <john.mcnamara@intel.com>,
Marko Kovacevic <marko.kovacevic@intel.com>
Cc: <dev@dpdk.org>
Subject: [dpdk-dev] [PATCH] net/octeontx2: add support VLAN insert and strip action
Date: Mon, 24 Aug 2020 12:34:24 +0530 [thread overview]
Message-ID: <20200824070424.252040-1-kirankumark@marvell.com> (raw)
From: Kiran Kumar K <kirankumark@marvell.com>
Adding support for RTE Flow VLAN insert and strip actions for octeontx2
device.
Signed-off-by: Kiran Kumar K <kirankumark@marvell.com>
---
doc/guides/nics/octeontx2.rst | 68 +++++++------
drivers/net/octeontx2/otx2_flow.c | 126 ++++++++++++++++++++++++
drivers/net/octeontx2/otx2_flow.h | 6 ++
drivers/net/octeontx2/otx2_flow_parse.c | 76 ++++++++++++--
drivers/net/octeontx2/otx2_flow_utils.c | 15 +--
5 files changed, 237 insertions(+), 54 deletions(-)
diff --git a/doc/guides/nics/octeontx2.rst b/doc/guides/nics/octeontx2.rst
index bb591a8b7..f3be79bbb 100644
--- a/doc/guides/nics/octeontx2.rst
+++ b/doc/guides/nics/octeontx2.rst
@@ -385,38 +385,46 @@ Actions:
.. table:: Ingress action types
- +----+--------------------------------+
- | # | Action Type |
- +====+================================+
- | 1 | RTE_FLOW_ACTION_TYPE_VOID |
- +----+--------------------------------+
- | 2 | RTE_FLOW_ACTION_TYPE_MARK |
- +----+--------------------------------+
- | 3 | RTE_FLOW_ACTION_TYPE_FLAG |
- +----+--------------------------------+
- | 4 | RTE_FLOW_ACTION_TYPE_COUNT |
- +----+--------------------------------+
- | 5 | RTE_FLOW_ACTION_TYPE_DROP |
- +----+--------------------------------+
- | 6 | RTE_FLOW_ACTION_TYPE_QUEUE |
- +----+--------------------------------+
- | 7 | RTE_FLOW_ACTION_TYPE_RSS |
- +----+--------------------------------+
- | 8 | RTE_FLOW_ACTION_TYPE_SECURITY |
- +----+--------------------------------+
- | 9 | RTE_FLOW_ACTION_TYPE_PF |
- +----+--------------------------------+
- | 10 | RTE_FLOW_ACTION_TYPE_VF |
- +----+--------------------------------+
+ +----+-----------------------------------------+
+ | # | Action Type |
+ +====+=========================================+
+ | 1 | RTE_FLOW_ACTION_TYPE_VOID |
+ +----+-----------------------------------------+
+ | 2 | RTE_FLOW_ACTION_TYPE_MARK |
+ +----+-----------------------------------------+
+ | 3 | RTE_FLOW_ACTION_TYPE_FLAG |
+ +----+-----------------------------------------+
+ | 4 | RTE_FLOW_ACTION_TYPE_COUNT |
+ +----+-----------------------------------------+
+ | 5 | RTE_FLOW_ACTION_TYPE_DROP |
+ +----+-----------------------------------------+
+ | 6 | RTE_FLOW_ACTION_TYPE_QUEUE |
+ +----+-----------------------------------------+
+ | 7 | RTE_FLOW_ACTION_TYPE_RSS |
+ +----+-----------------------------------------+
+ | 8 | RTE_FLOW_ACTION_TYPE_SECURITY |
+ +----+-----------------------------------------+
+ | 9 | RTE_FLOW_ACTION_TYPE_PF |
+ +----+-----------------------------------------+
+ | 10 | RTE_FLOW_ACTION_TYPE_VF |
+ +----+-----------------------------------------+
+ | 11 | RTE_FLOW_ACTION_TYPE_OF_POP_VLAN |
+ +----+-----------------------------------------+
.. _table_octeontx2_supported_egress_action_types:
.. table:: Egress action types
- +----+--------------------------------+
- | # | Action Type |
- +====+================================+
- | 1 | RTE_FLOW_ACTION_TYPE_COUNT |
- +----+--------------------------------+
- | 2 | RTE_FLOW_ACTION_TYPE_DROP |
- +----+--------------------------------+
+ +----+-----------------------------------------+
+ | # | Action Type |
+ +====+=========================================+
+ | 1 | RTE_FLOW_ACTION_TYPE_COUNT |
+ +----+-----------------------------------------+
+ | 2 | RTE_FLOW_ACTION_TYPE_DROP |
+ +----+-----------------------------------------+
+ | 3 | RTE_FLOW_ACTION_TYPE_OF_PUSH_VLAN |
+ +----+-----------------------------------------+
+ | 4 | RTE_FLOW_ACTION_TYPE_OF_SET_VLAN_VID |
+ +----+-----------------------------------------+
+ | 5 | RTE_FLOW_ACTION_TYPE_OF_SET_VLAN_PCP |
+ +----+-----------------------------------------+
diff --git a/drivers/net/octeontx2/otx2_flow.c b/drivers/net/octeontx2/otx2_flow.c
index 13a76e441..9b32b2acc 100644
--- a/drivers/net/octeontx2/otx2_flow.c
+++ b/drivers/net/octeontx2/otx2_flow.c
@@ -6,6 +6,8 @@
#include "otx2_ethdev_sec.h"
#include "otx2_flow.h"
+enum flow_vtag_cfg_dir { VTAG_TX, VTAG_RX };
+
int
otx2_flow_free_all_resources(struct otx2_eth_dev *hw)
{
@@ -460,6 +462,109 @@ otx2_flow_validate(struct rte_eth_dev *dev,
&parse_state);
}
+static int
+flow_program_vtag_action(struct rte_eth_dev *eth_dev,
+ const struct rte_flow_action actions[],
+ struct rte_flow *flow)
+{
+ uint16_t vlan_id = 0, vlan_ethtype = RTE_ETHER_TYPE_VLAN;
+ struct otx2_eth_dev *dev = eth_dev->data->dev_private;
+ union {
+ uint64_t reg;
+ struct nix_tx_vtag_action_s act;
+ } tx_vtag_action;
+ struct otx2_mbox *mbox = dev->mbox;
+ struct nix_vtag_config *vtag_cfg;
+ struct nix_vtag_config_rsp *rsp;
+ bool vlan_insert_action = false;
+ uint64_t rx_vtag_action = 0;
+ uint8_t vlan_pcp = 0;
+ int rc;
+
+ for (; actions->type != RTE_FLOW_ACTION_TYPE_END; actions++) {
+ if (actions->type == RTE_FLOW_ACTION_TYPE_OF_POP_VLAN) {
+ if (dev->npc_flow.vtag_actions == 1) {
+ vtag_cfg =
+ otx2_mbox_alloc_msg_nix_vtag_cfg(mbox);
+ vtag_cfg->cfg_type = VTAG_RX;
+ vtag_cfg->rx.strip_vtag = 1;
+ /* Always capture */
+ vtag_cfg->rx.capture_vtag = 1;
+ vtag_cfg->vtag_size = NIX_VTAGSIZE_T4;
+ vtag_cfg->rx.vtag_type = 0;
+
+ rc = otx2_mbox_process(mbox);
+ if (rc)
+ return rc;
+ }
+
+ rx_vtag_action |= (NIX_RX_VTAGACTION_VTAG_VALID << 15);
+ rx_vtag_action |= (NPC_LID_LB << 8);
+ rx_vtag_action |= NIX_RX_VTAGACTION_VTAG0_RELPTR;
+ flow->vtag_action = rx_vtag_action;
+ } else if (actions->type ==
+ RTE_FLOW_ACTION_TYPE_OF_SET_VLAN_VID) {
+ const struct rte_flow_action_of_set_vlan_vid *vtag =
+ (const struct rte_flow_action_of_set_vlan_vid *)
+ actions->conf;
+ vlan_id = rte_be_to_cpu_16(vtag->vlan_vid);
+ if (vlan_id > 0xfff) {
+ otx2_err("Invalid vlan_id for set vlan action");
+ return -EINVAL;
+ }
+ vlan_insert_action = true;
+ } else if (actions->type == RTE_FLOW_ACTION_TYPE_OF_PUSH_VLAN) {
+ const struct rte_flow_action_of_push_vlan *ethtype =
+ (const struct rte_flow_action_of_push_vlan *)
+ actions->conf;
+ vlan_ethtype = rte_be_to_cpu_16(ethtype->ethertype);
+ if (vlan_ethtype != RTE_ETHER_TYPE_VLAN &&
+ vlan_ethtype != RTE_ETHER_TYPE_QINQ) {
+ otx2_err("Invalid ethtype specified for push"
+ " vlan action");
+ return -EINVAL;
+ }
+ vlan_insert_action = true;
+ } else if (actions->type ==
+ RTE_FLOW_ACTION_TYPE_OF_SET_VLAN_PCP) {
+ const struct rte_flow_action_of_set_vlan_pcp *pcp =
+ (const struct rte_flow_action_of_set_vlan_pcp *)
+ actions->conf;
+ vlan_pcp = pcp->vlan_pcp;
+ if (vlan_pcp > 0x7) {
+ otx2_err("Invalid PCP value for pcp action");
+ return -EINVAL;
+ }
+ vlan_insert_action = true;
+ }
+ }
+
+ if (vlan_insert_action) {
+ vtag_cfg = otx2_mbox_alloc_msg_nix_vtag_cfg(mbox);
+ vtag_cfg->cfg_type = VTAG_TX;
+ vtag_cfg->vtag_size = NIX_VTAGSIZE_T4;
+ vtag_cfg->tx.vtag0 =
+ ((vlan_ethtype << 16) | (vlan_pcp << 13) | vlan_id);
+ vtag_cfg->tx.cfg_vtag0 = 1;
+ rc = otx2_mbox_process_msg(mbox, (void *)&rsp);
+ if (rc)
+ return rc;
+
+ tx_vtag_action.reg = 0;
+ tx_vtag_action.act.vtag0_def = rsp->vtag0_idx;
+ if (tx_vtag_action.act.vtag0_def < 0) {
+ otx2_err("Failed to config TX VTAG action");
+ return -EINVAL;
+ }
+ tx_vtag_action.act.vtag0_lid = NPC_LID_LA;
+ tx_vtag_action.act.vtag0_op = NIX_TX_VTAGOP_INSERT;
+ tx_vtag_action.act.vtag0_relptr =
+ NIX_TX_VTAGACTION_VTAG0_RELPTR;
+ flow->vtag_action = tx_vtag_action.reg;
+ }
+ return 0;
+}
+
static struct rte_flow *
otx2_flow_create(struct rte_eth_dev *dev,
const struct rte_flow_attr *attr,
@@ -489,6 +594,15 @@ otx2_flow_create(struct rte_eth_dev *dev,
if (rc != 0)
goto err_exit;
+ rc = flow_program_vtag_action(dev, actions, flow);
+ if (rc != 0) {
+ rte_flow_error_set(error, EIO,
+ RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
+ NULL,
+ "Failed to program vlan action");
+ goto err_exit;
+ }
+
rc = flow_program_npc(&parse_state, mbox, &hw->npc_flow);
if (rc != 0) {
rte_flow_error_set(error, EIO,
@@ -561,6 +675,17 @@ otx2_flow_destroy(struct rte_eth_dev *dev,
}
}
+ if (flow->nix_intf == OTX2_INTF_RX && flow->vtag_action) {
+ npc->vtag_actions--;
+ if (npc->vtag_actions == 0) {
+ if (hw->vlan_info.strip_on == 0) {
+ hw->rx_offload_flags &=
+ ~NIX_RX_OFFLOAD_VLAN_STRIP_F;
+ otx2_eth_set_rx_function(dev);
+ }
+ }
+ }
+
rc = flow_free_rss_action(dev, flow);
if (rc != 0) {
rte_flow_error_set(error, EIO,
@@ -849,6 +974,7 @@ otx2_flow_init(struct otx2_eth_dev *hw)
}
rte_atomic32_init(&npc->mark_actions);
+ npc->vtag_actions = 0;
npc->mcam_entries = NPC_MCAM_TOT_ENTRIES >> npc->keyw[NPC_MCAM_RX];
/* Free, free_rev, live and live_rev entries */
diff --git a/drivers/net/octeontx2/otx2_flow.h b/drivers/net/octeontx2/otx2_flow.h
index df78f41d3..1f118c408 100644
--- a/drivers/net/octeontx2/otx2_flow.h
+++ b/drivers/net/octeontx2/otx2_flow.h
@@ -55,6 +55,10 @@ enum {
#define OTX2_FLOW_ACT_COUNT (1 << 7)
#define OTX2_FLOW_ACT_PF (1 << 8)
#define OTX2_FLOW_ACT_VF (1 << 9)
+#define OTX2_FLOW_ACT_VLAN_STRIP (1 << 10)
+#define OTX2_FLOW_ACT_VLAN_INSERT (1 << 11)
+#define OTX2_FLOW_ACT_VLAN_ETHTYPE_INSERT (1 << 12)
+#define OTX2_FLOW_ACT_VLAN_PCP_INSERT (1 << 13)
/* terminating actions */
#define OTX2_FLOW_ACT_TERM (OTX2_FLOW_ACT_DROP | \
@@ -154,6 +158,7 @@ struct rte_flow {
uint64_t mcam_data[OTX2_MAX_MCAM_WIDTH_DWORDS];
uint64_t mcam_mask[OTX2_MAX_MCAM_WIDTH_DWORDS];
uint64_t npc_action;
+ uint64_t vtag_action;
TAILQ_ENTRY(rte_flow) next;
};
@@ -162,6 +167,7 @@ TAILQ_HEAD(otx2_flow_list, rte_flow);
/* Accessed from ethdev private - otx2_eth_dev */
struct otx2_npc_flow_info {
rte_atomic32_t mark_actions;
+ uint32_t vtag_actions;
uint32_t keyx_supp_nmask[NPC_MAX_INTF];/* nibble mask */
uint32_t keyx_len[NPC_MAX_INTF]; /* per intf key len in bits */
uint32_t datax_len[NPC_MAX_INTF]; /* per intf data len in bits */
diff --git a/drivers/net/octeontx2/otx2_flow_parse.c b/drivers/net/octeontx2/otx2_flow_parse.c
index 2d9a5857c..b7ff5fd03 100644
--- a/drivers/net/octeontx2/otx2_flow_parse.c
+++ b/drivers/net/octeontx2/otx2_flow_parse.c
@@ -813,6 +813,7 @@ otx2_flow_parse_actions(struct rte_eth_dev *dev,
const struct rte_flow_action_mark *act_mark;
const struct rte_flow_action_queue *act_q;
const struct rte_flow_action_vf *vf_act;
+ bool vlan_insert_action = false;
const char *errmsg = NULL;
int sel_act, req_act = 0;
uint16_t pf_func, vf_id;
@@ -932,6 +933,18 @@ otx2_flow_parse_actions(struct rte_eth_dev *dev,
req_act |= OTX2_FLOW_ACT_SEC;
rq = 0;
break;
+ case RTE_FLOW_ACTION_TYPE_OF_SET_VLAN_VID:
+ req_act |= OTX2_FLOW_ACT_VLAN_INSERT;
+ break;
+ case RTE_FLOW_ACTION_TYPE_OF_POP_VLAN:
+ req_act |= OTX2_FLOW_ACT_VLAN_STRIP;
+ break;
+ case RTE_FLOW_ACTION_TYPE_OF_PUSH_VLAN:
+ req_act |= OTX2_FLOW_ACT_VLAN_ETHTYPE_INSERT;
+ break;
+ case RTE_FLOW_ACTION_TYPE_OF_SET_VLAN_PCP:
+ req_act |= OTX2_FLOW_ACT_VLAN_PCP_INSERT;
+ break;
default:
errmsg = "Unsupported action specified";
errcode = ENOTSUP;
@@ -939,20 +952,46 @@ otx2_flow_parse_actions(struct rte_eth_dev *dev,
}
}
+ if (req_act &
+ (OTX2_FLOW_ACT_VLAN_INSERT | OTX2_FLOW_ACT_VLAN_ETHTYPE_INSERT |
+ OTX2_FLOW_ACT_VLAN_PCP_INSERT))
+ vlan_insert_action = true;
+
+ if ((req_act &
+ (OTX2_FLOW_ACT_VLAN_INSERT | OTX2_FLOW_ACT_VLAN_ETHTYPE_INSERT |
+ OTX2_FLOW_ACT_VLAN_PCP_INSERT)) ==
+ OTX2_FLOW_ACT_VLAN_PCP_INSERT) {
+ errmsg = " PCP insert action can't be supported alone";
+ errcode = ENOTSUP;
+ goto err_exit;
+ }
+
+ /* Both STRIP and INSERT actions are not supported */
+ if (vlan_insert_action && (req_act & OTX2_FLOW_ACT_VLAN_STRIP)) {
+ errmsg = "Both VLAN insert and strip actions not supported"
+ " together";
+ errcode = ENOTSUP;
+ goto err_exit;
+ }
+
/* Check if actions specified are compatible */
if (attr->egress) {
- /* Only DROP/COUNT is supported */
- if (!(req_act & OTX2_FLOW_ACT_DROP)) {
- errmsg = "DROP is required action for egress";
- errcode = EINVAL;
- goto err_exit;
- } else if (req_act & ~(OTX2_FLOW_ACT_DROP |
- OTX2_FLOW_ACT_COUNT)) {
- errmsg = "Unsupported action specified";
+ if (req_act & OTX2_FLOW_ACT_VLAN_STRIP) {
+ errmsg = "VLAN pop action is not supported on Egress";
errcode = ENOTSUP;
goto err_exit;
}
- flow->npc_action = NIX_TX_ACTIONOP_DROP;
+
+ if (req_act & OTX2_FLOW_ACT_DROP) {
+ flow->npc_action = NIX_TX_ACTIONOP_DROP;
+ } else if ((req_act & OTX2_FLOW_ACT_COUNT) ||
+ vlan_insert_action) {
+ flow->npc_action = NIX_TX_ACTIONOP_UCAST_DEFAULT;
+ } else {
+ errmsg = "Unsupported action for egress";
+ errcode = EINVAL;
+ goto err_exit;
+ }
goto set_pf_func;
}
@@ -985,8 +1024,20 @@ otx2_flow_parse_actions(struct rte_eth_dev *dev,
goto err_exit;
}
+ if (vlan_insert_action) {
+ errmsg = "VLAN push/Insert action is not supported on Ingress";
+ errcode = ENOTSUP;
+ goto err_exit;
+ }
+
+ if (req_act & OTX2_FLOW_ACT_VLAN_STRIP)
+ npc->vtag_actions++;
+
+ /* Only VLAN action is provided */
+ if (req_act == OTX2_FLOW_ACT_VLAN_STRIP)
+ flow->npc_action = NIX_RX_ACTIONOP_UCAST;
/* Set NIX_RX_ACTIONOP */
- if (req_act & (OTX2_FLOW_ACT_PF | OTX2_FLOW_ACT_VF)) {
+ else if (req_act & (OTX2_FLOW_ACT_PF | OTX2_FLOW_ACT_VF)) {
flow->npc_action = NIX_RX_ACTIONOP_UCAST;
if (req_act & OTX2_FLOW_ACT_QUEUE)
flow->npc_action |= (uint64_t)rq << 20;
@@ -1032,6 +1083,11 @@ otx2_flow_parse_actions(struct rte_eth_dev *dev,
otx2_eth_set_rx_function(dev);
}
+ if (npc->vtag_actions == 1) {
+ hw->rx_offload_flags |= NIX_RX_OFFLOAD_VLAN_STRIP_F;
+ otx2_eth_set_rx_function(dev);
+ }
+
set_pf_func:
/* Ideally AF must ensure that correct pf_func is set */
flow->npc_action |= (uint64_t)pf_func << 4;
diff --git a/drivers/net/octeontx2/otx2_flow_utils.c b/drivers/net/octeontx2/otx2_flow_utils.c
index 14625c9ad..6215a542f 100644
--- a/drivers/net/octeontx2/otx2_flow_utils.c
+++ b/drivers/net/octeontx2/otx2_flow_utils.c
@@ -916,20 +916,7 @@ otx2_flow_mcam_alloc_and_write(struct rte_flow *flow, struct otx2_mbox *mbox,
(flow->nix_intf == OTX2_INTF_RX) ? NPC_MCAM_RX : NPC_MCAM_TX;
req->enable_entry = 1;
req->entry_data.action = flow->npc_action;
-
- /*
- * DPDK sets vtag action on per interface basis, not
- * per flow basis. It is a matter of how we decide to support
- * this pmd specific behavior. There are two ways:
- * 1. Inherit the vtag action from the one configured
- * for this interface. This can be read from the
- * vtag_action configured for default mcam entry of
- * this pf_func.
- * 2. Do not support vtag action with rte_flow.
- *
- * Second approach is used now.
- */
- req->entry_data.vtag_action = 0ULL;
+ req->entry_data.vtag_action = flow->vtag_action;
for (idx = 0; idx < OTX2_MAX_MCAM_WIDTH_DWORDS; idx++) {
req->entry_data.kw[idx] = flow->mcam_data[idx];
--
2.25.1
next reply other threads:[~2020-08-24 7:04 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-08-24 7:04 kirankumark [this message]
2020-09-29 9:51 ` Jerin Jacob
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200824070424.252040-1-kirankumark@marvell.com \
--to=kirankumark@marvell.com \
--cc=dev@dpdk.org \
--cc=jerinj@marvell.com \
--cc=john.mcnamara@intel.com \
--cc=marko.kovacevic@intel.com \
--cc=ndabilpuram@marvell.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).