From: Ferruh Yigit <ferruh.yigit@amd.com>
To: Jesna K E <jesna.k.e@amd.com>, dev@dpdk.org
Cc: Selwin.Sebastian@amd.com
Subject: Re: [PATCH v1 2/2] net/axgbe: move offloads to Rx/Tx queue setup
Date: Wed, 21 Dec 2022 16:27:23 +0000 [thread overview]
Message-ID: <a6e62458-20aa-ad6f-6469-7d612f6f514c@amd.com> (raw)
In-Reply-To: <20221221025202.31733-2-jesna.k.e@amd.com>
On 12/21/2022 2:52 AM, Jesna K E wrote:
> For Multiprocess dpdk applications retrieving the offload parameter
> for receive packet directly from struct eth_dev
> giving segmentation fault since rxmode.offloads from eth_dev was
> null duirng recv_pkts. So retrieved offload from rx/tx_queue_setup()
>
I don't see any issue to store offload values in queue_setup functions,
and it can be more performant to access if from queue struct,
but I am not clear why it cause segfault when it is in data path, when
burst functions start to run device should be configured alredy, finding
root case helps us to figure out if anything is missing, hence:
Can you please give more details on why/how seg fault happens?
> Signed-off-by: Jesna K E <jesna.k.e@amd.com>
> ---
> drivers/net/axgbe/axgbe_rxtx.c | 10 ++++------
> drivers/net/axgbe/axgbe_rxtx.h | 4 ++--
> 2 files changed, 6 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/net/axgbe/axgbe_rxtx.c b/drivers/net/axgbe/axgbe_rxtx.c
> index 9b283bd9d0..5dc4fe7369 100644
> --- a/drivers/net/axgbe/axgbe_rxtx.c
> +++ b/drivers/net/axgbe/axgbe_rxtx.c
> @@ -86,6 +86,7 @@ int axgbe_dev_rx_queue_setup(struct rte_eth_dev *dev, uint16_t queue_idx,
> if (rxq->free_thresh > rxq->nb_desc)
> rxq->free_thresh = rxq->nb_desc >> 3;
>
> + rxq->offloads = rx_conf->offloads | dev->data->dev_conf.rxmode.offloads;
> /* Allocate RX ring hardware descriptors */
> size = rxq->nb_desc * sizeof(union axgbe_rx_desc);
> dma = rte_eth_dma_zone_reserve(dev, "rx_ring", queue_idx, size, 128,
> @@ -211,7 +212,6 @@ axgbe_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
> unsigned int err, etlt;
> uint32_t error_status;
> uint16_t idx, pidx, pkt_len;
> - uint64_t offloads;
>
> idx = AXGBE_GET_DESC_IDX(rxq, rxq->cur);
> while (nb_rx < nb_pkts) {
> @@ -278,14 +278,13 @@ axgbe_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
> mbuf->hash.rss = rte_le_to_cpu_32(desc->write.desc1);
> etlt = AXGMAC_GET_BITS_LE(desc->write.desc3,
> RX_NORMAL_DESC3, ETLT);
> - offloads = rxq->pdata->eth_dev->data->dev_conf.rxmode.offloads;
> if (!err || !etlt) {
> if (etlt == RX_CVLAN_TAG_PRESENT) {
> mbuf->ol_flags |= RTE_MBUF_F_RX_VLAN;
> mbuf->vlan_tci =
> AXGMAC_GET_BITS_LE(desc->write.desc0,
> RX_NORMAL_DESC0, OVT);
> - if (offloads & RTE_ETH_RX_OFFLOAD_VLAN_STRIP)
> + if (rxq->offloads & RTE_ETH_RX_OFFLOAD_VLAN_STRIP)
> mbuf->ol_flags |= RTE_MBUF_F_RX_VLAN_STRIPPED;
> else
> mbuf->ol_flags &= ~RTE_MBUF_F_RX_VLAN_STRIPPED;
> @@ -345,7 +344,6 @@ uint16_t eth_axgbe_recv_scattered_pkts(void *rx_queue,
> unsigned int err = 0, etlt;
> uint32_t error_status = 0;
> uint16_t idx, pidx, data_len = 0, pkt_len = 0;
> - uint64_t offloads;
> bool eop = 0;
>
> idx = AXGBE_GET_DESC_IDX(rxq, rxq->cur);
> @@ -441,14 +439,13 @@ uint16_t eth_axgbe_recv_scattered_pkts(void *rx_queue,
> rte_le_to_cpu_32(desc->write.desc1);
> etlt = AXGMAC_GET_BITS_LE(desc->write.desc3,
> RX_NORMAL_DESC3, ETLT);
> - offloads = rxq->pdata->eth_dev->data->dev_conf.rxmode.offloads;
> if (!err || !etlt) {
> if (etlt == RX_CVLAN_TAG_PRESENT) {
> first_seg->ol_flags |= RTE_MBUF_F_RX_VLAN;
> first_seg->vlan_tci =
> AXGMAC_GET_BITS_LE(desc->write.desc0,
> RX_NORMAL_DESC0, OVT);
> - if (offloads & RTE_ETH_RX_OFFLOAD_VLAN_STRIP)
> + if (rxq->offloads & RTE_ETH_RX_OFFLOAD_VLAN_STRIP)
> first_seg->ol_flags |=
> RTE_MBUF_F_RX_VLAN_STRIPPED;
> else
> @@ -606,6 +603,7 @@ int axgbe_dev_tx_queue_setup(struct rte_eth_dev *dev, uint16_t queue_idx,
> txq->desc = tz->addr;
> txq->queue_id = queue_idx;
> txq->port_id = dev->data->port_id;
> + txq->offloads = offloads;
> txq->dma_regs = (void *)((uint8_t *)pdata->xgmac_regs + DMA_CH_BASE +
> (DMA_CH_INC * txq->queue_id));
> txq->dma_tail_reg = (volatile uint32_t *)((uint8_t *)txq->dma_regs +
> diff --git a/drivers/net/axgbe/axgbe_rxtx.h b/drivers/net/axgbe/axgbe_rxtx.h
> index d5660f5c4b..56e7c7aadd 100644
> --- a/drivers/net/axgbe/axgbe_rxtx.h
> +++ b/drivers/net/axgbe/axgbe_rxtx.h
> @@ -100,7 +100,7 @@ struct axgbe_rx_queue {
> uint64_t rx_mbuf_alloc_failed;
> /* Number of mbufs allocated from pool*/
> uint64_t mbuf_alloc;
> -
> + uint64_t offloads; /**< Rx offloads with RTE_ETH_RX_OFFLOAD_**/
> } __rte_cache_aligned;
>
> /*Tx descriptor format */
> @@ -149,7 +149,7 @@ struct axgbe_tx_queue {
> uint64_t pkts;
> uint64_t bytes;
> uint64_t errors;
> -
> + uint64_t offloads; /**< Tx offload flags of RTE_ETH_TX_OFFLOAD_* */
> } __rte_cache_aligned;
>
> /*Queue related APIs */
next prev parent reply other threads:[~2022-12-21 16:27 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-12-21 2:52 [PATCH v1 1/2] net/axgbe: add multi-process support Jesna K E
2022-12-21 2:52 ` [PATCH v1 2/2] net/axgbe: move offloads to Rx/Tx queue setup Jesna K E
2022-12-21 16:27 ` Ferruh Yigit [this message]
2022-12-21 15:37 ` [PATCH v1 1/2] net/axgbe: add multi-process support Ferruh Yigit
2022-12-23 10:44 ` Ferruh Yigit
2022-12-23 10:47 ` Ferruh Yigit
-- strict thread matches above, loose matches on Subject: below --
2023-01-04 7:42 Jesna K E
2023-01-04 7:42 ` [PATCH v1 2/2] net/axgbe: move offloads to Rx/Tx queue setup Jesna K E
2022-12-21 2:35 [PATCH v1 1/2] net/axgbe: add multi-process support Jesna K E
2022-12-21 2:35 ` [PATCH v1 2/2] net/axgbe: move offloads to Rx/Tx queue setup Jesna K E
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=a6e62458-20aa-ad6f-6469-7d612f6f514c@amd.com \
--to=ferruh.yigit@amd.com \
--cc=Selwin.Sebastian@amd.com \
--cc=dev@dpdk.org \
--cc=jesna.k.e@amd.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).