From: "Gavin Hu (Arm Technology China)" <Gavin.Hu@arm.com>
To: Di ChenxuX <chenxux.di@intel.com>, "dev@dpdk.org" <dev@dpdk.org>
Cc: "qiming.yang@intel.com" <qiming.yang@intel.com>, nd <nd@arm.com>
Subject: Re: [dpdk-dev] [PATCH 2/4] net/i40e: cleanup Tx buffers
Date: Thu, 26 Sep 2019 10:36:17 +0000 [thread overview]
Message-ID: <VI1PR08MB5376E78F8DD9FD5DFECEC91A8F860@VI1PR08MB5376.eurprd08.prod.outlook.com> (raw)
In-Reply-To: <20190926092933.79683-3-chenxux.di@intel.com>
> -----Original Message-----
> From: dev <dev-bounces@dpdk.org> On Behalf Of Di ChenxuX
> Sent: Thursday, September 26, 2019 5:30 PM
> To: dev@dpdk.org
> Cc: qiming.yang@intel.com; Di ChenxuX <chenxux.di@intel.com>
> Subject: [dpdk-dev] [PATCH 2/4] net/i40e: cleanup Tx buffers
>
> Add support to the i40e driver for the API rte_eth_tx_done_cleanup
> to force free consumed buffers on Tx ring.
>
> Signed-off-by: Di ChenxuX <chenxux.di@intel.com>
> ---
> drivers/net/i40e/i40e_ethdev.c | 1 +
> drivers/net/i40e/i40e_ethdev_vf.c | 1 +
> drivers/net/i40e/i40e_rxtx.c | 42 +++++++++++++++++++++++++++++++
> drivers/net/i40e/i40e_rxtx.h | 1 +
> 4 files changed, 45 insertions(+)
>
> diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
> index 4e40b7ab5..cf35fb5da 100644
> --- a/drivers/net/i40e/i40e_ethdev.c
> +++ b/drivers/net/i40e/i40e_ethdev.c
> @@ -509,6 +509,7 @@ static const struct eth_dev_ops i40e_eth_dev_ops =
> {
> .mac_addr_set = i40e_set_default_mac_addr,
> .mtu_set = i40e_dev_mtu_set,
> .tm_ops_get = i40e_tm_ops_get,
> + .tx_done_cleanup = i40e_tx_done_cleanup,
> };
>
> /* store statistics names and its offset in stats structure */
> diff --git a/drivers/net/i40e/i40e_ethdev_vf.c
> b/drivers/net/i40e/i40e_ethdev_vf.c
> index c77b30c54..b462a9d8c 100644
> --- a/drivers/net/i40e/i40e_ethdev_vf.c
> +++ b/drivers/net/i40e/i40e_ethdev_vf.c
> @@ -215,6 +215,7 @@ static const struct eth_dev_ops
> i40evf_eth_dev_ops = {
> .rss_hash_conf_get = i40evf_dev_rss_hash_conf_get,
> .mtu_set = i40evf_dev_mtu_set,
> .mac_addr_set = i40evf_set_default_mac_addr,
> + .tx_done_cleanup = i40e_tx_done_cleanup,
> };
>
> /*
> diff --git a/drivers/net/i40e/i40e_rxtx.c b/drivers/net/i40e/i40e_rxtx.c
> index 692c3bab4..29793a1b7 100644
> --- a/drivers/net/i40e/i40e_rxtx.c
> +++ b/drivers/net/i40e/i40e_rxtx.c
> @@ -1416,6 +1416,8 @@ i40e_xmit_pkts_vec(void *tx_queue, struct
> rte_mbuf **tx_pkts,
> break;
> }
>
> + //eth_i40e_tx_done_cleanup(tx_queue,5);
> +
> return nb_tx;
> }
>
> @@ -2467,6 +2469,46 @@ i40e_tx_queue_release_mbufs(struct
> i40e_tx_queue *txq)
> }
> }
>
> +int i40e_tx_done_cleanup(void *txq, uint32_t free_cnt)
> +{
> + struct i40e_tx_queue *q = (struct i40e_tx_queue *)txq;
> + struct i40e_tx_entry *sw_ring;
> + uint16_t tx_id; /* Current segment being processed. */
> + uint16_t tx_cleaned;
> +
> + int count = 0;
uint32_t is better, then you will not get trouble into a negative value.
> +
> + if (q == NULL)
> + return -ENODEV;
> +
> + sw_ring = q->sw_ring;
> + tx_cleaned = q->last_desc_cleaned;
> + tx_id = sw_ring[q->last_desc_cleaned].next_id;
> + if ((q->tx_ring[tx_id].cmd_type_offset_bsz &
> +
> rte_cpu_to_le_64(I40E_TXD_QW1_DTYPE_MASK)) !=
> +
> rte_cpu_to_le_64(I40E_TX_DESC_DTYPE_DESC_DONE))
> + return 0;
> +
> + do {
> + if (sw_ring[tx_id].mbuf == NULL)
> + break;
> +
> + rte_pktmbuf_free_seg(sw_ring[tx_id].mbuf);
> + sw_ring[tx_id].mbuf = NULL;
> + sw_ring[tx_id].last_id = tx_id;
> +
> + /* Move to next segemnt. */
> + tx_cleaned = tx_id;
> + tx_id = sw_ring[tx_id].next_id;
> + count++;
> + } while (count != (int)free_cnt);
This force conversation is not needed if "count" declared as uint32_t above.
/Gavin
> +
> + q->nb_tx_free += (uint16_t)count;
> + q->last_desc_cleaned = tx_cleaned;
> +
> + return count;
> +}
> +
> void
> i40e_reset_tx_queue(struct i40e_tx_queue *txq)
> {
> diff --git a/drivers/net/i40e/i40e_rxtx.h b/drivers/net/i40e/i40e_rxtx.h
> index 3fc619af9..1a70eda2c 100644
> --- a/drivers/net/i40e/i40e_rxtx.h
> +++ b/drivers/net/i40e/i40e_rxtx.h
> @@ -204,6 +204,7 @@ void i40e_dev_free_queues(struct rte_eth_dev
> *dev);
> void i40e_reset_rx_queue(struct i40e_rx_queue *rxq);
> void i40e_reset_tx_queue(struct i40e_tx_queue *txq);
> void i40e_tx_queue_release_mbufs(struct i40e_tx_queue *txq);
> +int i40e_tx_done_cleanup(void *txq, uint32_t free_cnt);
> int i40e_alloc_rx_queue_mbufs(struct i40e_rx_queue *rxq);
> void i40e_rx_queue_release_mbufs(struct i40e_rx_queue *rxq);
>
> --
> 2.17.1
next prev parent reply other threads:[~2019-09-26 10:36 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-26 9:29 [dpdk-dev] [PATCH 0/4] drivers/net: " Di ChenxuX
2019-09-26 9:29 ` [dpdk-dev] [PATCH 1/4] net/fm10k: " Di ChenxuX
2019-09-26 10:37 ` Gavin Hu (Arm Technology China)
2019-09-26 9:29 ` [dpdk-dev] [PATCH 2/4] net/i40e: " Di ChenxuX
2019-09-26 10:36 ` Gavin Hu (Arm Technology China) [this message]
2019-09-26 9:29 ` [dpdk-dev] [PATCH 3/4] net/ice: " Di ChenxuX
2019-09-26 9:29 ` [dpdk-dev] [PATCH 4/4] net/ixgbe: " Di ChenxuX
2019-12-03 5:51 [dpdk-dev] [PATCH 0/4] drivers/net: " Chenxu Di
2019-12-03 5:51 ` [dpdk-dev] [PATCH 2/4] net/i40e: " Chenxu Di
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=VI1PR08MB5376E78F8DD9FD5DFECEC91A8F860@VI1PR08MB5376.eurprd08.prod.outlook.com \
--to=gavin.hu@arm.com \
--cc=chenxux.di@intel.com \
--cc=dev@dpdk.org \
--cc=nd@arm.com \
--cc=qiming.yang@intel.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).