DPDK patches and discussions
 help / color / mirror / Atom feed
From: xiangxia.m.yue@gmail.com
To: viacheslavo@mellanox.com
Cc: dev@dpdk.org, Tonghao Zhang <xiangxia.m.yue@gmail.com>, stable@dpdk.org
Subject: [dpdk-dev] [PATCH dpdk-dev v2] net/mlx5: check the reg available for metadata action
Date: Sun, 24 May 2020 19:06:41 +0800
Message-ID: <20200524110641.72147-1-xiangxia.m.yue@gmail.com> (raw)
In-Reply-To: <20200520013328.98838-1-xiangxia.m.yue@gmail.com>

From: Tonghao Zhang <xiangxia.m.yue@gmail.com>

If user don't set the dv_xmeta_en to 1 or 2,
in the flow_dv_convert_action_set_meta function:
* flow_dv_get_metadata_reg may return the REG_NONE,
  when MLX5_METADATA_FDB enabled for metadata set
  action.
* reg_to_field(REG_NONE) return MLX5_MODI_OUT_NONE
  which is invalid.

The rdma-core calltrace:
  dr_action_create_modify_action
  dr_actions_convert_modify_header
  dr_action_modify_sw_to_hw
  dr_action_modify_sw_to_hw_set
  dr_ste_get_modify_hdr_hw_field

sw_field [MLX5_MODI_OUT_NONE 4095]
should not > ste_ctx->modify_hdr_field_arr_sz [92]

As doc[1] says:
| dv_xmeta_en 0, this is default value, defines the legacy mode,
| the MARK and META related actions and items operate only within
| NIC Tx and NIC Rx steering domains, no MARK and META information
| crosses the domain boundaries.

This patch add check for that case to warn that not supported.

[1] - http://doc.dpdk.org/guides-20.02/nics/mlx5.html
Fixes: fcc8d2f716fd ("net/mlx5: extend flow metadata support")
Cc: stable@dpdk.org

Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
---
 drivers/net/mlx5/mlx5_flow_dv.c | 31 +++++++++++++++++++------------
 1 file changed, 19 insertions(+), 12 deletions(-)

diff --git a/drivers/net/mlx5/mlx5_flow_dv.c b/drivers/net/mlx5/mlx5_flow_dv.c
index e4818319507c..c77835270d60 100644
--- a/drivers/net/mlx5/mlx5_flow_dv.c
+++ b/drivers/net/mlx5/mlx5_flow_dv.c
@@ -1251,6 +1251,7 @@ flow_dv_convert_action_set_meta
 
 	if (reg < 0)
 		return reg;
+	MLX5_ASSERT(reg != REG_NONE);
 	/*
 	 * In datapath code there is no endianness
 	 * coversions for perfromance reasons, all
@@ -1449,7 +1450,6 @@ flow_dv_validate_item_meta(struct rte_eth_dev *dev __rte_unused,
 			   struct rte_flow_error *error)
 {
 	struct mlx5_priv *priv = dev->data->dev_private;
-	struct mlx5_dev_config *config = &priv->config;
 	const struct rte_flow_item_meta *spec = item->spec;
 	const struct rte_flow_item_meta *mask = item->mask;
 	struct rte_flow_item_meta nic_mask = {
@@ -1463,23 +1463,25 @@ flow_dv_validate_item_meta(struct rte_eth_dev *dev __rte_unused,
 					  RTE_FLOW_ERROR_TYPE_ITEM_SPEC,
 					  item->spec,
 					  "data cannot be empty");
-	if (config->dv_xmeta_en != MLX5_XMETA_MODE_LEGACY) {
-		if (!mlx5_flow_ext_mreg_supported(dev))
-			return rte_flow_error_set(error, ENOTSUP,
+	if (!mlx5_flow_ext_mreg_supported(dev))
+		return rte_flow_error_set(error, ENOTSUP,
 					  RTE_FLOW_ERROR_TYPE_ITEM, item,
 					  "extended metadata register"
 					  " isn't supported");
-		reg = flow_dv_get_metadata_reg(dev, attr, error);
-		if (reg < 0)
-			return reg;
-		if (reg == REG_B)
-			return rte_flow_error_set(error, ENOTSUP,
+	reg = flow_dv_get_metadata_reg(dev, attr, error);
+	if (reg < 0)
+		return reg;
+	if (reg == REG_NONE)
+		return rte_flow_error_set(error, ENOTSUP,
+					  RTE_FLOW_ERROR_TYPE_ITEM, item,
+					  "unavailable metadata register");
+	if (reg == REG_B)
+		return rte_flow_error_set(error, ENOTSUP,
 					  RTE_FLOW_ERROR_TYPE_ITEM, item,
 					  "match on reg_b "
 					  "isn't supported");
-		if (reg != REG_A)
-			nic_mask.data = priv->sh->dv_meta_mask;
-	}
+	if (reg != REG_A)
+		nic_mask.data = priv->sh->dv_meta_mask;
 	if (!mask)
 		mask = &rte_flow_item_meta_mask;
 	if (!mask->data)
@@ -2244,6 +2246,11 @@ flow_dv_validate_action_set_meta(struct rte_eth_dev *dev,
 	reg = flow_dv_get_metadata_reg(dev, attr, error);
 	if (reg < 0)
 		return reg;
+	if (reg == REG_NONE)
+		return rte_flow_error_set(error, ENOTSUP,
+					  RTE_FLOW_ERROR_TYPE_ACTION,
+					  action, "unavailable "
+					  "metadata register");
 	if (reg != REG_A && reg != REG_B) {
 		struct mlx5_priv *priv = dev->data->dev_private;
 
-- 
2.26.1


  parent reply	other threads:[~2020-05-24 11:07 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-20  1:33 [dpdk-dev] [PATCH dpdk-dev] " xiangxia.m.yue
2020-05-22 17:17 ` Slava Ovsiienko
2020-05-24 10:47   ` Tonghao Zhang
2020-05-27  7:43     ` Tonghao Zhang
2020-05-24 11:06 ` xiangxia.m.yue [this message]
2020-08-05 17:09   ` [dpdk-dev] [PATCH dpdk-dev v2] " Slava Ovsiienko
2020-11-20 14:48   ` [dpdk-dev] [PATCH v3] " Viacheslav Ovsiienko
2020-11-22 13:00     ` Raslan Darawsheh

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=20200524110641.72147-1-xiangxia.m.yue@gmail.com \
    --to=xiangxia.m.yue@gmail.com \
    --cc=dev@dpdk.org \
    --cc=stable@dpdk.org \
    --cc=viacheslavo@mellanox.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

DPDK patches and discussions

This inbox may be cloned and mirrored by anyone:

	git clone --mirror https://inbox.dpdk.org/dev/0 dev/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 dev dev/ https://inbox.dpdk.org/dev \
		dev@dpdk.org
	public-inbox-index dev

Example config snippet for mirrors.
Newsgroup available over NNTP:
	nntp://inbox.dpdk.org/inbox.dpdk.dev


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git