From: Tal Shnaiderman <talshn@nvidia.com>
To: <dev@dpdk.org>
Cc: <thomas@monjalon.net>, <matan@nvidia.com>, <rasland@nvidia.com>,
<asafp@nvidia.com>, <viacheslavo@nvidia.com>, <eilong@nvidia.com>,
<kcollins@nvidia.com>, <idanhac@nvidia.com>, <stable@dpdk.org>
Subject: [dpdk-stable] [PATCH v2 01/13] net/mlx5: fix software parsing support query
Date: Tue, 12 Oct 2021 15:45:42 +0300 [thread overview]
Message-ID: <20211012124554.21296-2-talshn@nvidia.com> (raw)
In-Reply-To: <20211012124554.21296-1-talshn@nvidia.com>
Currently, the PMD decides if the software parsing
offload can enable outer IPv4 checksum and tunneled
TSO support by checking config->hw_csum and config->tso
respectively.
This is incorrect, the right way is to check the following
flags returned by the mlx5dv_query_device function:
MLX5DV_SW_PARSING - check general swp support.
MLX5DV_SW_PARSING_CSUM - check swp checksum support.
MLX5DV_SW_PARSING_LSO - check swp LSO/TSO support.
The fix enables the offloads according to the correct
flags returned by the kernel.
Fixes: e46821e9fcdc60 ("net/mlx5: separate generic tunnel TSO from the standard one")
Cc: stable@dpdk.org
Signed-off-by: Tal Shnaiderman <talshn@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Tested-by: Idan Hackmon <idanhac@nvidia.com>
---
drivers/net/mlx5/linux/mlx5_os.c | 3 ++-
drivers/net/mlx5/linux/mlx5_os.h | 12 ++++++++++++
drivers/net/mlx5/mlx5.h | 2 +-
drivers/net/mlx5/mlx5_txq.c | 15 +++++++++------
4 files changed, 24 insertions(+), 8 deletions(-)
diff --git a/drivers/net/mlx5/linux/mlx5_os.c b/drivers/net/mlx5/linux/mlx5_os.c
index 3746057673..a6542629c7 100644
--- a/drivers/net/mlx5/linux/mlx5_os.c
+++ b/drivers/net/mlx5/linux/mlx5_os.c
@@ -1112,7 +1112,8 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev,
swp = dv_attr.sw_parsing_caps.sw_parsing_offloads;
DRV_LOG(DEBUG, "SWP support: %u", swp);
#endif
- config->swp = !!swp;
+ config->swp = swp & (MLX5_SW_PARSING_CAP | MLX5_SW_PARSING_CSUM_CAP |
+ MLX5_SW_PARSING_TSO_CAP);
#ifdef HAVE_IBV_DEVICE_STRIDING_RQ_SUPPORT
if (dv_attr.comp_mask & MLX5DV_CONTEXT_MASK_STRIDING_RQ) {
struct mlx5dv_striding_rq_caps mprq_caps =
diff --git a/drivers/net/mlx5/linux/mlx5_os.h b/drivers/net/mlx5/linux/mlx5_os.h
index 2991d37df2..da036edb72 100644
--- a/drivers/net/mlx5/linux/mlx5_os.h
+++ b/drivers/net/mlx5/linux/mlx5_os.h
@@ -21,4 +21,16 @@ enum {
int mlx5_auxiliary_get_ifindex(const char *sf_name);
+
+enum mlx5_sw_parsing_offloads {
+#ifdef HAVE_IBV_MLX5_MOD_SWP
+ MLX5_SW_PARSING_CAP = MLX5DV_SW_PARSING,
+ MLX5_SW_PARSING_CSUM_CAP = MLX5DV_SW_PARSING_CSUM,
+ MLX5_SW_PARSING_TSO_CAP = MLX5DV_SW_PARSING_LSO,
+#else
+ MLX5_SW_PARSING_CAP = 0,
+ MLX5_SW_PARSING_CSUM_CAP = 0,
+ MLX5_SW_PARSING_TSO_CAP = 0,
+#endif
+};
#endif /* RTE_PMD_MLX5_OS_H_ */
diff --git a/drivers/net/mlx5/mlx5.h b/drivers/net/mlx5/mlx5.h
index 3581414b78..edb4f26d42 100644
--- a/drivers/net/mlx5/mlx5.h
+++ b/drivers/net/mlx5/mlx5.h
@@ -260,7 +260,7 @@ struct mlx5_dev_config {
unsigned int dv_xmeta_en:2; /* Enable extensive flow metadata. */
unsigned int lacp_by_user:1;
/* Enable user to manage LACP traffic. */
- unsigned int swp:1; /* Tx generic tunnel checksum and TSO offload. */
+ unsigned int swp:3; /* Tx generic tunnel checksum and TSO offload. */
unsigned int devx:1; /* Whether devx interface is available or not. */
unsigned int dest_tir:1; /* Whether advanced DR API is available. */
unsigned int reclaim_mode:2; /* Memory reclaim mode. */
diff --git a/drivers/net/mlx5/mlx5_txq.c b/drivers/net/mlx5/mlx5_txq.c
index eb4d34ca55..8dca2b7f79 100644
--- a/drivers/net/mlx5/mlx5_txq.c
+++ b/drivers/net/mlx5/mlx5_txq.c
@@ -111,9 +111,9 @@ mlx5_get_tx_port_offloads(struct rte_eth_dev *dev)
if (config->tx_pp)
offloads |= DEV_TX_OFFLOAD_SEND_ON_TIMESTAMP;
if (config->swp) {
- if (config->hw_csum)
+ if (config->swp & MLX5_SW_PARSING_CSUM_CAP)
offloads |= DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM;
- if (config->tso)
+ if (config->swp & MLX5_SW_PARSING_TSO_CAP)
offloads |= (DEV_TX_OFFLOAD_IP_TNL_TSO |
DEV_TX_OFFLOAD_UDP_TNL_TSO);
}
@@ -979,10 +979,13 @@ txq_set_params(struct mlx5_txq_ctrl *txq_ctrl)
txq_ctrl->txq.tso_en = 1;
}
txq_ctrl->txq.tunnel_en = config->tunnel_en | config->swp;
- txq_ctrl->txq.swp_en = ((DEV_TX_OFFLOAD_IP_TNL_TSO |
- DEV_TX_OFFLOAD_UDP_TNL_TSO |
- DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM) &
- txq_ctrl->txq.offloads) && config->swp;
+ txq_ctrl->txq.swp_en = (((DEV_TX_OFFLOAD_IP_TNL_TSO |
+ DEV_TX_OFFLOAD_UDP_TNL_TSO) &
+ txq_ctrl->txq.offloads) && (config->swp &
+ MLX5_SW_PARSING_TSO_CAP)) |
+ ((DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM &
+ txq_ctrl->txq.offloads) && (config->swp &
+ MLX5_SW_PARSING_CSUM_CAP));
}
/**
--
2.16.1.windows.4
next parent reply other threads:[~2021-10-12 12:46 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20211012124554.21296-1-talshn@nvidia.com>
2021-10-12 12:45 ` Tal Shnaiderman [this message]
2021-10-12 12:45 ` [dpdk-stable] [PATCH v2 04/13] net/mlx5: fix tunneling " Tal Shnaiderman
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=20211012124554.21296-2-talshn@nvidia.com \
--to=talshn@nvidia.com \
--cc=asafp@nvidia.com \
--cc=dev@dpdk.org \
--cc=eilong@nvidia.com \
--cc=idanhac@nvidia.com \
--cc=kcollins@nvidia.com \
--cc=matan@nvidia.com \
--cc=rasland@nvidia.com \
--cc=stable@dpdk.org \
--cc=thomas@monjalon.net \
--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).