DPDK patches and discussions
 help / color / mirror / Atom feed
From: Jerin Jacob <jerinjacobk@gmail.com>
To: skoteshwar@marvell.com
Cc: Nithin Dabilpuram <ndabilpuram@marvell.com>,
	Kiran Kumar K <kirankumark@marvell.com>,
	 Sunil Kumar Kori <skori@marvell.com>,
	dev@dpdk.org
Subject: Re: [PATCH] net/cnxk: flush SQ before configuring MTU
Date: Thu, 15 Jun 2023 14:20:45 +0530	[thread overview]
Message-ID: <CALBAE1PcDZahNHncO3FXMjM294=gWHiL8-t+pwmOfirrdGqtEA@mail.gmail.com> (raw)
In-Reply-To: <1686805463-19231-1-git-send-email-skoteshwar@marvell.com>

On Thu, Jun 15, 2023 at 10:34 AM <skoteshwar@marvell.com> wrote:
>
> From: Satha Rao <skoteshwar@marvell.com>
>
> When try to configure MTU for lower value causes run time failure
> due to old bigger packets enqueued. To avoid error interrupts better
> to flush the all SQs of this port before configuring new MTU.

Added
    Fixes: 8589ec212e80 ("net/cnxk: support MTU set")
    Cc: stable@dpdk.org


Applied to dpdk-next-net-mrvl/for-next-net. Thanks

>
> Signed-off-by: Satha Rao <skoteshwar@marvell.com>
> ---
>  drivers/net/cnxk/cnxk_ethdev.h     |  1 +
>  drivers/net/cnxk/cnxk_ethdev_ops.c | 47 ++++++++++++++++++++++++++++++++++++++
>  2 files changed, 48 insertions(+)
>
> diff --git a/drivers/net/cnxk/cnxk_ethdev.h b/drivers/net/cnxk/cnxk_ethdev.h
> index e280d6c..45460ae 100644
> --- a/drivers/net/cnxk/cnxk_ethdev.h
> +++ b/drivers/net/cnxk/cnxk_ethdev.h
> @@ -446,6 +446,7 @@ int cnxk_nix_probe(struct rte_pci_driver *pci_drv,
>                    struct rte_pci_device *pci_dev);
>  int cnxk_nix_remove(struct rte_pci_device *pci_dev);
>  int cnxk_nix_mtu_set(struct rte_eth_dev *eth_dev, uint16_t mtu);
> +int cnxk_nix_sq_flush(struct rte_eth_dev *eth_dev);
>  int cnxk_nix_mc_addr_list_configure(struct rte_eth_dev *eth_dev,
>                                     struct rte_ether_addr *mc_addr_set,
>                                     uint32_t nb_mc_addr);
> diff --git a/drivers/net/cnxk/cnxk_ethdev_ops.c b/drivers/net/cnxk/cnxk_ethdev_ops.c
> index bce6d59..da5ee19 100644
> --- a/drivers/net/cnxk/cnxk_ethdev_ops.c
> +++ b/drivers/net/cnxk/cnxk_ethdev_ops.c
> @@ -496,6 +496,44 @@
>  }
>
>  int
> +cnxk_nix_sq_flush(struct rte_eth_dev *eth_dev)
> +{
> +       struct cnxk_eth_dev *dev = cnxk_eth_pmd_priv(eth_dev);
> +       struct rte_eth_dev_data *data = eth_dev->data;
> +       int i, rc = 0;
> +
> +       /* Flush all tx queues */
> +       for (i = 0; i < eth_dev->data->nb_tx_queues; i++) {
> +               struct roc_nix_sq *sq = &dev->sqs[i];
> +
> +               if (eth_dev->data->tx_queues[i] == NULL)
> +                       continue;
> +
> +               rc = roc_nix_tm_sq_aura_fc(sq, false);
> +               if (rc) {
> +                       plt_err("Failed to disable sqb aura fc, rc=%d", rc);
> +                       goto exit;
> +               }
> +
> +               /* Wait for sq entries to be flushed */
> +               rc = roc_nix_tm_sq_flush_spin(sq);
> +               if (rc) {
> +                       plt_err("Failed to drain sq, rc=%d\n", rc);
> +                       goto exit;
> +               }
> +               if (data->tx_queue_state[i] == RTE_ETH_QUEUE_STATE_STARTED) {
> +                       rc = roc_nix_tm_sq_aura_fc(sq, true);
> +                       if (rc) {
> +                               plt_err("Failed to enable sq aura fc, txq=%u, rc=%d", i, rc);
> +                               goto exit;
> +                       }
> +               }
> +       }
> +exit:
> +       return rc;
> +}
> +
> +int
>  cnxk_nix_mtu_set(struct rte_eth_dev *eth_dev, uint16_t mtu)
>  {
>         uint32_t old_frame_size, frame_size = mtu + CNXK_NIX_L2_OVERHEAD;
> @@ -538,6 +576,15 @@
>                 goto exit;
>         }
>
> +       /* if new MTU was smaller than old one, then flush all SQs before MTU change */
> +       if (old_frame_size > frame_size) {
> +               if (data->dev_started) {
> +                       plt_err("Reducing MTU is not supported when device started");
> +                       goto exit;
> +               }
> +               cnxk_nix_sq_flush(eth_dev);
> +       }
> +
>         frame_size -= RTE_ETHER_CRC_LEN;
>
>         /* Update mtu on Tx */
> --
> 1.8.3.1
>

      reply	other threads:[~2023-06-15  8:51 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-06-15  5:04 skoteshwar
2023-06-15  8:50 ` 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='CALBAE1PcDZahNHncO3FXMjM294=gWHiL8-t+pwmOfirrdGqtEA@mail.gmail.com' \
    --to=jerinjacobk@gmail.com \
    --cc=dev@dpdk.org \
    --cc=kirankumark@marvell.com \
    --cc=ndabilpuram@marvell.com \
    --cc=skori@marvell.com \
    --cc=skoteshwar@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).