DPDK patches and discussions
 help / color / mirror / Atom feed
From: Somnath Kotur <somnath.kotur@broadcom.com>
To: dev@dpdk.org
Cc: ferruh.yigit@intel.com
Subject: [dpdk-dev] [PATCH 12/17] net/bnxt: fix segfault after removing and adding the slaves
Date: Thu, 24 Oct 2019 13:14:27 +0530	[thread overview]
Message-ID: <20191024074432.30705-13-somnath.kotur@broadcom.com> (raw)
In-Reply-To: <20191024074432.30705-1-somnath.kotur@broadcom.com>

From: Santoshkumar Karanappa Rastapur <santosh.rastapur@broadcom.com>

On removing the slave interface, slave_remove in bonding module calls
_rte_eth_dev_reset which in turn frees both Tx and Rx queues.
1. segfault is seen after removing/adding the slave interface and starting
bond interface.
In this below path, when mtu is set for the slave interface, queues are
not created yet and driver reference to queue[0] causes the segfault.
slave_configure:
	rte_eth_dev_set_mtu
	rte_eth_dev_configure
	rte_eth_rx_queue_setup

2. segfault is seen on starting the port after removing from bond device.
This is a testpmd bug where in, on starting the port, testpmd is supposed
to recreate the queues before starting the port.

Fixed these by adding check for queues created before accessing them.

Signed-off-by: Santoshkumar Karanappa Rastapur <santosh.rastapur@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
---
 drivers/net/bnxt/bnxt.h        |  1 +
 drivers/net/bnxt/bnxt_ethdev.c | 19 ++++++++++++++++---
 drivers/net/bnxt/bnxt_rxq.c    |  4 ++++
 3 files changed, 21 insertions(+), 3 deletions(-)

diff --git a/drivers/net/bnxt/bnxt.h b/drivers/net/bnxt/bnxt.h
index 975bcf9..ffc37ea 100644
--- a/drivers/net/bnxt/bnxt.h
+++ b/drivers/net/bnxt/bnxt.h
@@ -639,6 +639,7 @@ struct bnxt {
 	struct bnxt_error_recovery_info *recovery_info;
 };
 
+int bnxt_mtu_set_op(struct rte_eth_dev *eth_dev, uint16_t new_mtu);
 int bnxt_link_update_op(struct rte_eth_dev *eth_dev, int wait_to_complete);
 int bnxt_rcv_msg_from_vf(struct bnxt *bp, uint16_t vf_id, void *msg);
 int is_bnxt_in_error(struct bnxt *bp);
diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index e24fd41..4187247 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -121,7 +121,6 @@ static const struct rte_pci_id bnxt_pci_id_map[] = {
 
 static int bnxt_vlan_offload_set_op(struct rte_eth_dev *dev, int mask);
 static void bnxt_print_link_info(struct rte_eth_dev *eth_dev);
-static int bnxt_mtu_set_op(struct rte_eth_dev *eth_dev, uint16_t new_mtu);
 static int bnxt_dev_uninit(struct rte_eth_dev *eth_dev);
 static int bnxt_init_resources(struct bnxt *bp, bool reconfig_dev);
 static int bnxt_uninit_resources(struct bnxt *bp, bool reconfig_dev);
@@ -831,6 +830,11 @@ static int bnxt_dev_start_op(struct rte_eth_dev *eth_dev)
 	int vlan_mask = 0;
 	int rc;
 
+	if (!eth_dev->data->nb_tx_queues || !eth_dev->data->nb_rx_queues) {
+		PMD_DRV_LOG(ERR, "Queues are not configured yet!\n");
+		return -EINVAL;
+	}
+
 	if (bp->rx_cp_nr_rings > RTE_ETHDEV_QUEUE_STAT_CNTRS) {
 		PMD_DRV_LOG(ERR,
 			"RxQ cnt %d > CONFIG_RTE_ETHDEV_QUEUE_STAT_CNTRS %d\n",
@@ -2126,7 +2130,7 @@ bnxt_txq_info_get_op(struct rte_eth_dev *dev, uint16_t queue_id,
 	qinfo->conf.tx_deferred_start = txq->tx_deferred_start;
 }
 
-static int bnxt_mtu_set_op(struct rte_eth_dev *eth_dev, uint16_t new_mtu)
+int bnxt_mtu_set_op(struct rte_eth_dev *eth_dev, uint16_t new_mtu)
 {
 	struct bnxt *bp = eth_dev->data->dev_private;
 	uint32_t new_pkt_size;
@@ -2137,6 +2141,10 @@ static int bnxt_mtu_set_op(struct rte_eth_dev *eth_dev, uint16_t new_mtu)
 	if (rc)
 		return rc;
 
+	/* Exit if receive queues are not configured yet */
+	if (!eth_dev->data->nb_rx_queues)
+		return rc;
+
 	new_pkt_size = new_mtu + RTE_ETHER_HDR_LEN + RTE_ETHER_CRC_LEN +
 		       VLAN_TAG_SIZE * BNXT_NUM_VLANS;
 
@@ -2167,7 +2175,9 @@ static int bnxt_mtu_set_op(struct rte_eth_dev *eth_dev, uint16_t new_mtu)
 		bp->flags &= ~BNXT_FLAG_JUMBO;
 	}
 
-	eth_dev->data->dev_conf.rxmode.max_rx_pkt_len = new_pkt_size;
+	/* Is there a change in mtu setting? */
+	if (eth_dev->data->dev_conf.rxmode.max_rx_pkt_len == new_pkt_size)
+		return rc;
 
 	for (i = 0; i < bp->nr_vnics; i++) {
 		struct bnxt_vnic_info *vnic = &bp->vnic_info[i];
@@ -2188,6 +2198,9 @@ static int bnxt_mtu_set_op(struct rte_eth_dev *eth_dev, uint16_t new_mtu)
 		}
 	}
 
+	if (!rc)
+		eth_dev->data->dev_conf.rxmode.max_rx_pkt_len = new_pkt_size;
+
 	PMD_DRV_LOG(INFO, "New MTU is %d\n", new_mtu);
 
 	return rc;
diff --git a/drivers/net/bnxt/bnxt_rxq.c b/drivers/net/bnxt/bnxt_rxq.c
index e1ed360..d55adc3 100644
--- a/drivers/net/bnxt/bnxt_rxq.c
+++ b/drivers/net/bnxt/bnxt_rxq.c
@@ -366,6 +366,10 @@ int bnxt_rx_queue_setup_op(struct rte_eth_dev *eth_dev,
 	eth_dev->data->rx_queue_state[queue_idx] = queue_state;
 	rte_spinlock_init(&rxq->lock);
 
+	/* Configure mtu if it is different from what was configured before */
+	if (!queue_idx)
+		bnxt_mtu_set_op(eth_dev, eth_dev->data->mtu);
+
 out:
 	return rc;
 }
-- 
2.10.1.613.g2cc2e70


  parent reply	other threads:[~2019-10-24  7:54 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-24  7:44 [dpdk-dev] [PATCH v2 00/17] bnxt patchset with bug fixes Somnath Kotur
2019-10-24  7:44 ` [dpdk-dev] [PATCH 01/17] net/bnxt: fix unused value in free filter mem Somnath Kotur
2019-10-24  7:44 ` [dpdk-dev] [PATCH 02/17] net/bnxt: fix ping failure with MTU change Somnath Kotur
2019-10-24  7:44 ` [dpdk-dev] [PATCH v2 03/17] net/bnxt: fix to set MAX RSS contexts for maximal reservation Somnath Kotur
2019-10-24  7:44 ` [dpdk-dev] [PATCH 04/17] net/bnxt: fix to write the correct MTU to FW Somnath Kotur
2019-10-24  7:44 ` [dpdk-dev] [PATCH 05/17] net/bnxt: fix RSS configuration failure for thor-based controllers Somnath Kotur
2019-10-24  7:44 ` [dpdk-dev] [PATCH 06/17] net/bnxt: update HWRM API to version 1.10.1.6 Somnath Kotur
2019-10-25 13:22   ` Ferruh Yigit
2019-10-24  7:44 ` [dpdk-dev] [PATCH 07/17] net/bnxt: use mac index, while checking for default mac Somnath Kotur
2019-10-24  7:44 ` [dpdk-dev] [PATCH 08/17] net/bnxt: fix a possible race between start and interrupt handler Somnath Kotur
2019-10-24  7:44 ` [dpdk-dev] [PATCH 09/17] net/bnxt: fix to use HWRM_CFA_L2_FILTER as the filter_type for non-ntuple flows Somnath Kotur
2019-10-24  7:44 ` [dpdk-dev] [PATCH 10/17] net/bnxt: expose some missing counters in port statistics Somnath Kotur
2019-10-24  7:44 ` [dpdk-dev] [PATCH 11/17] net/bnxt: fix to check for zero mac address while parsing flow arguments Somnath Kotur
2019-10-24  7:44 ` Somnath Kotur [this message]
2019-10-24  7:44 ` [dpdk-dev] [PATCH 13/17] net/bnxt: add more checks for fw reset Somnath Kotur
2019-10-24  7:44 ` [dpdk-dev] [PATCH 14/17] net/bnxt: add PCI IDs for 57500 series NPAR devices Somnath Kotur
2019-10-24  7:44 ` [dpdk-dev] [PATCH 15/17] net/bnxt: add a wrapper function for thor device check Somnath Kotur
2019-10-24  7:44 ` [dpdk-dev] [PATCH 16/17] net/bnxt: fix a possible memory leak Somnath Kotur
2019-10-24  7:44 ` [dpdk-dev] [PATCH 17/17] net/bnxt: call bnxt_uninit_locks from bnxt_uninit_resources Somnath Kotur
2019-10-24 18:40 ` [dpdk-dev] [PATCH v2 00/17] bnxt patchset with bug fixes Ajit Khaparde

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=20191024074432.30705-13-somnath.kotur@broadcom.com \
    --to=somnath.kotur@broadcom.com \
    --cc=dev@dpdk.org \
    --cc=ferruh.yigit@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).