From: Dariusz Sosnowski <dsosnowski@nvidia.com>
To: Viacheslav Ovsiienko <viacheslavo@nvidia.com>,
Bing Zhao <bingz@nvidia.com>, Ori Kam <orika@nvidia.com>,
Suanming Mou <suanmingm@nvidia.com>,
Matan Azrad <matan@nvidia.com>
Cc: <dev@dpdk.org>, Raslan Darawsheh <rasland@nvidia.com>, <stable@dpdk.org>
Subject: [PATCH] net/mlx5: fix NAT64 register selection
Date: Wed, 12 Mar 2025 10:51:08 +0100 [thread overview]
Message-ID: <20250312095108.105978-1-dsosnowski@nvidia.com> (raw)
PMD statically assumed that REG_C_6 is always available for use
with NAT64 HW flow action.
This led to PMD configuration errors on FW versions which do not expose
that specific register.
This patch fixes that by adding a check for REG_C_6 against FW
capabilities, when registers for NAT64 are selected.
Also, if not enough registers are available, PMD will not attempt to
create NAT64 actions.
Fixes: 7a26bfec06a4 ("net/mlx5: fetch available registers for NAT64")
Cc: stable@dpdk.org
Signed-off-by: Dariusz Sosnowski <dsosnowski@nvidia.com>
Acked-by: Bing Zhao <bingz@nvidia.com>
---
drivers/net/mlx5/mlx5.c | 16 +++++++++++-----
drivers/net/mlx5/mlx5_flow_hw.c | 22 ++++++++++++++++++++--
2 files changed, 31 insertions(+), 7 deletions(-)
diff --git a/drivers/net/mlx5/mlx5.c b/drivers/net/mlx5/mlx5.c
index 0f49cb5e5b..e175f81031 100644
--- a/drivers/net/mlx5/mlx5.c
+++ b/drivers/net/mlx5/mlx5.c
@@ -1624,7 +1624,8 @@ mlx5_init_hws_flow_tags_registers(struct mlx5_dev_ctx_shared *sh)
{
struct mlx5_dev_registers *reg = &sh->registers;
uint32_t meta_mode = sh->config.dv_xmeta_en;
- uint16_t masks = (uint16_t)sh->cdev->config.hca_attr.set_reg_c;
+ uint16_t reg_c_caps = (uint16_t)sh->cdev->config.hca_attr.set_reg_c;
+ uint16_t masks = reg_c_caps;
uint16_t unset = 0;
uint32_t i, j;
@@ -1648,11 +1649,16 @@ mlx5_init_hws_flow_tags_registers(struct mlx5_dev_ctx_shared *sh)
* Set the registers for NAT64 usage internally. REG_C_6 is always used.
* The other 2 registers will be fetched from right to left, at least 2
* tag registers should be available.
+ * If not enough registers are available or REG_C_6 is not supported by current FW,
+ * NAT64 action will not be supported.
*/
- MLX5_ASSERT(j >= (MLX5_FLOW_NAT64_REGS_MAX - 1));
- reg->nat64_regs[0] = REG_C_6;
- reg->nat64_regs[1] = reg->hw_avl_tags[j - 2];
- reg->nat64_regs[2] = reg->hw_avl_tags[j - 1];
+ if ((reg_c_caps & RTE_BIT32(mlx5_regc_index(REG_C_6))) &&
+ j >= MLX5_FLOW_NAT64_REGS_MAX - 1) {
+ MLX5_ASSERT(j >= (MLX5_FLOW_NAT64_REGS_MAX - 1));
+ reg->nat64_regs[0] = REG_C_6;
+ reg->nat64_regs[1] = reg->hw_avl_tags[j - 2];
+ reg->nat64_regs[2] = reg->hw_avl_tags[j - 1];
+ }
}
static void
diff --git a/drivers/net/mlx5/mlx5_flow_hw.c b/drivers/net/mlx5/mlx5_flow_hw.c
index 6254857301..20d38ce414 100644
--- a/drivers/net/mlx5/mlx5_flow_hw.c
+++ b/drivers/net/mlx5/mlx5_flow_hw.c
@@ -9656,6 +9656,19 @@ flow_hw_destroy_send_to_kernel_action(struct mlx5_priv *priv)
}
}
+static bool
+flow_hw_should_create_nat64_actions(struct mlx5_priv *priv)
+{
+ int i;
+
+ /* Check if all registers are available. */
+ for (i = 0; i < MLX5_FLOW_NAT64_REGS_MAX; ++i)
+ if (priv->sh->registers.nat64_regs[i] == REG_NON)
+ return false;
+
+ return true;
+}
+
static void
flow_hw_destroy_nat64_actions(struct mlx5_priv *priv)
{
@@ -12197,9 +12210,14 @@ __flow_hw_configure(struct rte_eth_dev *dev,
NULL, "Failed to VLAN actions.");
goto err;
}
- if (flow_hw_create_nat64_actions(priv, error))
+ if (flow_hw_should_create_nat64_actions(priv)) {
+ if (flow_hw_create_nat64_actions(priv, error))
+ goto err;
+ } else {
DRV_LOG(WARNING, "Cannot create NAT64 action on port %u, "
- "please check the FW version", dev->data->port_id);
+ "please check the FW version. NAT64 will not be supported.",
+ dev->data->port_id);
+ }
if (_queue_attr)
mlx5_free(_queue_attr);
if (port_attr->flags & RTE_FLOW_PORT_FLAG_STRICT_QUEUE)
--
2.39.5
next reply other threads:[~2025-03-12 9:51 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-12 9:51 Dariusz Sosnowski [this message]
2025-03-13 13:07 ` 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=20250312095108.105978-1-dsosnowski@nvidia.com \
--to=dsosnowski@nvidia.com \
--cc=bingz@nvidia.com \
--cc=dev@dpdk.org \
--cc=matan@nvidia.com \
--cc=orika@nvidia.com \
--cc=rasland@nvidia.com \
--cc=stable@dpdk.org \
--cc=suanmingm@nvidia.com \
--cc=viacheslavo@nvidia.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).