From: "Guo, Junfeng" <junfeng.guo@intel.com>
To: "Zeng, ZhichaoX" <zhichaox.zeng@intel.com>,
"dev@dpdk.org" <dev@dpdk.org>
Cc: "Yang, Qiming" <qiming.yang@intel.com>, "Su, Simei" <simei.su@intel.com>
Subject: RE: [PATCH] net/igc: fix segmentation fault in secondary dpdk-symmetric_mp
Date: Thu, 16 Jun 2022 02:49:58 +0000 [thread overview]
Message-ID: <DM6PR11MB3723F5B0200F78B0AA07F4CDE7AC9@DM6PR11MB3723.namprd11.prod.outlook.com> (raw)
In-Reply-To: <20220615073915.13309-1-zhichaox.zeng@intel.com>
> -----Original Message-----
> From: Zeng, ZhichaoX <zhichaox.zeng@intel.com>
> Sent: Wednesday, June 15, 2022 15:39
> To: dev@dpdk.org
> Cc: Yang, Qiming <qiming.yang@intel.com>; Zeng, ZhichaoX
> <zhichaox.zeng@intel.com>; Guo, Junfeng <junfeng.guo@intel.com>; Su,
> Simei <simei.su@intel.com>
> Subject: [PATCH] net/igc: fix segmentation fault in secondary dpdk-
> symmetric_mp
>
> From: Zhichao Zeng <zhichaox.zeng@intel.com>
>
> In the secondary dpdk-symmetric_mp process, the
> "smp_port_init" was skipped, which cause some function
> pointers not to be initialized, and a segmentation fault
> occurred when calling these function pointers.
>
> This patch assigns initial values to rx_pkt_burst,
> tx_pkt_burst and tx_pkt_prepare pointers to avoid
> calling null function pointer.
>
> Signed-off-by: Zhichao Zeng <zhichaox.zeng@intel.com>
> ---
> drivers/net/igc/igc_ethdev.c | 3 +++
> drivers/net/igc/igc_ethdev.h | 5 +++++
> drivers/net/igc/igc_txrx.c | 6 +++---
> 3 files changed, 11 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/net/igc/igc_ethdev.c b/drivers/net/igc/igc_ethdev.c
> index b9933b395d..dd4a028ea1 100644
> --- a/drivers/net/igc/igc_ethdev.c
> +++ b/drivers/net/igc/igc_ethdev.c
> @@ -1234,6 +1234,9 @@ eth_igc_dev_init(struct rte_eth_dev *dev)
> dev->rx_queue_count = eth_igc_rx_queue_count;
> dev->rx_descriptor_status = eth_igc_rx_descriptor_status;
> dev->tx_descriptor_status = eth_igc_tx_descriptor_status;
> + dev->rx_pkt_burst = &igc_recv_pkts;
> + dev->tx_pkt_burst = &igc_xmit_pkts;
> + dev->tx_pkt_prepare = ð_igc_prep_pkts;
Maybe the parameter & are not needed here? you can double check for this when build.
>
> /*
> * for secondary processes, we don't initialize any further as
> primary
> diff --git a/drivers/net/igc/igc_ethdev.h b/drivers/net/igc/igc_ethdev.h
> index f56cad79e9..2fa3d51bba 100644
> --- a/drivers/net/igc/igc_ethdev.h
> +++ b/drivers/net/igc/igc_ethdev.h
> @@ -263,6 +263,11 @@ struct igc_adapter {
> #define IGC_DEV_PRIVATE_FLOW_LIST(_dev) \
> (&((struct igc_adapter *)(_dev)->data->dev_private)->flow_list)
>
> +uint16_t igc_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
> uint16_t nb_pkts);
> +uint16_t igc_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
> uint16_t nb_pkts);
> +uint16_t eth_igc_prep_pkts(__rte_unused void *tx_queue, struct
> rte_mbuf **tx_pkts,
> +
> uint16_t nb_pkts);
> +
These functions are tx/rx related, it would be better to put them into igc_txrx.h.
> static inline void
> igc_read_reg_check_set_bits(struct igc_hw *hw, uint32_t reg, uint32_t
> bits)
> {
> diff --git a/drivers/net/igc/igc_txrx.c b/drivers/net/igc/igc_txrx.c
> index e48d5df11a..753ac8a411 100644
> --- a/drivers/net/igc/igc_txrx.c
> +++ b/drivers/net/igc/igc_txrx.c
> @@ -345,7 +345,7 @@ rx_desc_get_pkt_info(struct igc_rx_queue *rxq,
> struct rte_mbuf *rxm,
> rxm->packet_type = rx_desc_pkt_info_to_pkt_type(pkt_info);
> }
>
> -static uint16_t
> +uint16_t
> igc_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t
> nb_pkts)
> {
> struct igc_rx_queue * const rxq = rx_queue;
> @@ -1397,7 +1397,7 @@ eth_igc_rx_queue_setup(struct rte_eth_dev
> *dev,
> }
>
> /* prepare packets for transmit */
> -static uint16_t
> +uint16_t
> eth_igc_prep_pkts(__rte_unused void *tx_queue, struct rte_mbuf
> **tx_pkts,
> uint16_t nb_pkts)
> {
> @@ -1604,7 +1604,7 @@ tx_desc_cksum_flags_to_olinfo(uint64_t
> ol_flags)
> return tmp;
> }
>
> -static uint16_t
> +uint16_t
> igc_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, uint16_t
> nb_pkts)
> {
> struct igc_tx_queue * const txq = tx_queue;
> --
> 2.25.1
next prev parent reply other threads:[~2022-06-16 2:51 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-06-15 7:39 zhichaox.zeng
2022-06-16 2:49 ` Guo, Junfeng [this message]
2022-06-22 10:49 ` [PATCH v2] " zhichaox.zeng
2022-06-23 3:50 ` Zhang, Qi Z
2022-06-28 6:20 ` [PATCH v3] net/igc: move the initialization of data path into dev_init zhichaox.zeng
2022-06-28 7:00 ` Zhang, Qi Z
2022-06-30 11:03 ` [PATCH v4] net/igc: add support for secondary processes zhichaox.zeng
2022-06-30 11:47 ` Zhang, Qi Z
2022-07-01 3:13 ` Zeng, ZhichaoX
2022-07-01 3:47 ` Zhang, Qi Z
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=DM6PR11MB3723F5B0200F78B0AA07F4CDE7AC9@DM6PR11MB3723.namprd11.prod.outlook.com \
--to=junfeng.guo@intel.com \
--cc=dev@dpdk.org \
--cc=qiming.yang@intel.com \
--cc=simei.su@intel.com \
--cc=zhichaox.zeng@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).