From: Jerin Jacob <jerinjacobk@gmail.com>
To: Harman Kalra <hkalra@marvell.com>, Ferruh Yigit <ferruh.yigit@intel.com>
Cc: Jerin Jacob <jerinj@marvell.com>,
Nithin Dabilpuram <ndabilpuram@marvell.com>,
Kiran Kumar K <kirankumark@marvell.com>, dpdk-dev <dev@dpdk.org>,
dpdk stable <stable@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH] net/octeontx2: fix multi seg mode for jumbo packets
Date: Mon, 28 Sep 2020 10:51:18 +0530 [thread overview]
Message-ID: <CALBAE1NOArtZq56nRh=85fznXUd-6UPKg_VQCxuMLHqMtoUCew@mail.gmail.com> (raw)
In-Reply-To: <1597056945-23018-1-git-send-email-hkalra@marvell.com>
On Mon, Aug 10, 2020 at 4:26 PM Harman Kalra <hkalra@marvell.com> wrote:
>
> Scatter gather mode should get enabled:
> - If mbuf size is less than max received packet length.
> - MTU is reconfigured greater than mbuf size.
>
> Fixes: 0e2efd02db58 ("net/octeontx2: add MTU set operation")
> Cc: stable@dpdk.org
>
> Signed-off-by: Harman Kalra <hkalra@marvell.com>
Applied to dpdk-next-net-mrvl/master. Thanks
> ---
> drivers/net/octeontx2/otx2_ethdev.c | 30 +++++++++++++++++++++++++
> drivers/net/octeontx2/otx2_ethdev.h | 2 ++
> drivers/net/octeontx2/otx2_ethdev_ops.c | 9 +-------
> 3 files changed, 33 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/net/octeontx2/otx2_ethdev.c b/drivers/net/octeontx2/otx2_ethdev.c
> index 33b72bd4d..bee1e2a76 100644
> --- a/drivers/net/octeontx2/otx2_ethdev.c
> +++ b/drivers/net/octeontx2/otx2_ethdev.c
> @@ -657,6 +657,9 @@ otx2_nix_rx_queue_setup(struct rte_eth_dev *eth_dev, uint16_t rq,
> }
> }
>
> + /* Setup scatter mode if needed by jumbo */
> + otx2_nix_enable_mseg_on_jumbo(rxq);
> +
> return 0;
>
> free_rxq:
> @@ -878,6 +881,33 @@ nix_sqb_unlock(struct rte_mempool *mp)
> return 0;
> }
>
> +void
> +otx2_nix_enable_mseg_on_jumbo(struct otx2_eth_rxq *rxq)
> +{
> + struct rte_pktmbuf_pool_private *mbp_priv;
> + struct rte_eth_dev *eth_dev;
> + struct otx2_eth_dev *dev;
> + uint32_t buffsz;
> +
> + eth_dev = rxq->eth_dev;
> + dev = otx2_eth_pmd_priv(eth_dev);
> +
> + /* Get rx buffer size */
> + mbp_priv = rte_mempool_get_priv(rxq->pool);
> + buffsz = mbp_priv->mbuf_data_room_size - RTE_PKTMBUF_HEADROOM;
> +
> + if (eth_dev->data->dev_conf.rxmode.max_rx_pkt_len > buffsz) {
> + dev->rx_offloads |= DEV_RX_OFFLOAD_SCATTER;
> + dev->tx_offloads |= DEV_TX_OFFLOAD_MULTI_SEGS;
> +
> + /* Setting up the rx[tx]_offload_flags due to change
> + * in rx[tx]_offloads.
> + */
> + dev->rx_offload_flags |= nix_rx_offload_flags(eth_dev);
> + dev->tx_offload_flags |= nix_tx_offload_flags(eth_dev);
> + }
> +}
> +
> static int
> nix_sq_init(struct otx2_eth_txq *txq)
> {
> diff --git a/drivers/net/octeontx2/otx2_ethdev.h b/drivers/net/octeontx2/otx2_ethdev.h
> index e9efe52bb..11cdd8afd 100644
> --- a/drivers/net/octeontx2/otx2_ethdev.h
> +++ b/drivers/net/octeontx2/otx2_ethdev.h
> @@ -437,6 +437,8 @@ int otx2_nix_set_mc_addr_list(struct rte_eth_dev *eth_dev,
> /* MTU */
> int otx2_nix_mtu_set(struct rte_eth_dev *eth_dev, uint16_t mtu);
> int otx2_nix_recalc_mtu(struct rte_eth_dev *eth_dev);
> +void otx2_nix_enable_mseg_on_jumbo(struct otx2_eth_rxq *rxq);
> +
>
> /* Link */
> void otx2_nix_toggle_flag_link_cfg(struct otx2_eth_dev *dev, bool set);
> diff --git a/drivers/net/octeontx2/otx2_ethdev_ops.c b/drivers/net/octeontx2/otx2_ethdev_ops.c
> index faecf216c..b36d37b9f 100644
> --- a/drivers/net/octeontx2/otx2_ethdev_ops.c
> +++ b/drivers/net/octeontx2/otx2_ethdev_ops.c
> @@ -72,22 +72,15 @@ otx2_nix_mtu_set(struct rte_eth_dev *eth_dev, uint16_t mtu)
> int
> otx2_nix_recalc_mtu(struct rte_eth_dev *eth_dev)
> {
> - struct otx2_eth_dev *dev = otx2_eth_pmd_priv(eth_dev);
> struct rte_eth_dev_data *data = eth_dev->data;
> - struct rte_pktmbuf_pool_private *mbp_priv;
> struct otx2_eth_rxq *rxq;
> - uint32_t buffsz;
> uint16_t mtu;
> int rc;
>
> - /* Get rx buffer size */
> rxq = data->rx_queues[0];
> - mbp_priv = rte_mempool_get_priv(rxq->pool);
> - buffsz = mbp_priv->mbuf_data_room_size - RTE_PKTMBUF_HEADROOM;
>
> /* Setup scatter mode if needed by jumbo */
> - if (data->dev_conf.rxmode.max_rx_pkt_len > buffsz)
> - dev->rx_offloads |= DEV_RX_OFFLOAD_SCATTER;
> + otx2_nix_enable_mseg_on_jumbo(rxq);
>
> /* Setup MTU based on max_rx_pkt_len */
> mtu = data->dev_conf.rxmode.max_rx_pkt_len - NIX_L2_OVERHEAD;
> --
> 2.18.0
>
prev parent reply other threads:[~2020-09-28 5:21 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-08-10 10:55 Harman Kalra
2020-09-28 5:21 ` Jerin Jacob [this message]
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='CALBAE1NOArtZq56nRh=85fznXUd-6UPKg_VQCxuMLHqMtoUCew@mail.gmail.com' \
--to=jerinjacobk@gmail.com \
--cc=dev@dpdk.org \
--cc=ferruh.yigit@intel.com \
--cc=hkalra@marvell.com \
--cc=jerinj@marvell.com \
--cc=kirankumark@marvell.com \
--cc=ndabilpuram@marvell.com \
--cc=stable@dpdk.org \
/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).