| Message ID | 20241007132311.70593-1-anumula@chelsio.com (mailing list archive) |
|---|---|
| State | Accepted |
| Delegated to: | Jason Gunthorpe |
| Headers | show |
| Series | [for-rc] RDMA/cxgb4: fix RDMA_CM_EVENT_UNREACHABLE error for iWARP | expand |
On Mon, Oct 07, 2024 at 06:53:11PM +0530, Anumula Murali Mohan Reddy wrote: > ip_dev_find() always returns real net_device address, whether traffic is > running on a vlan or real device, if traffic is over vlan, filling > endpoint struture with real ndev and an attempt to send a connect > request will results in RDMA_CM_EVENT_UNREACHABLE error. > This patch fixes the issue by using vlan_dev_real_dev(). > > Fixes: 830662f6f032 ("RDMA/cxgb4: Add support for active and passive open connection with IPv6 address") > Signed-off-by: Anumula Murali Mohan Reddy <anumula@chelsio.com> > Signed-off-by: Potnuri Bharat Teja <bharat@chelsio.com> > --- > drivers/infiniband/hw/cxgb4/cm.c | 9 ++++----- > 1 file changed, 4 insertions(+), 5 deletions(-) Applied to for-rc, thanks Jason
diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index b3757c6a0457..8d753e6e0c71 100644 --- a/drivers/infiniband/hw/cxgb4/cm.c +++ b/drivers/infiniband/hw/cxgb4/cm.c @@ -2086,7 +2086,7 @@ static int import_ep(struct c4iw_ep *ep, int iptype, __u8 *peer_ip, err = -ENOMEM; if (n->dev->flags & IFF_LOOPBACK) { if (iptype == 4) - pdev = ip_dev_find(&init_net, *(__be32 *)peer_ip); + pdev = __ip_dev_find(&init_net, *(__be32 *)peer_ip, false); else if (IS_ENABLED(CONFIG_IPV6)) for_each_netdev(&init_net, pdev) { if (ipv6_chk_addr(&init_net, @@ -2101,12 +2101,12 @@ static int import_ep(struct c4iw_ep *ep, int iptype, __u8 *peer_ip, err = -ENODEV; goto out; } + if (is_vlan_dev(pdev)) + pdev = vlan_dev_real_dev(pdev); ep->l2t = cxgb4_l2t_get(cdev->rdev.lldi.l2t, n, pdev, rt_tos2priority(tos)); - if (!ep->l2t) { - dev_put(pdev); + if (!ep->l2t) goto out; - } ep->mtu = pdev->mtu; ep->tx_chan = cxgb4_port_chan(pdev); ep->smac_idx = ((struct port_info *)netdev_priv(pdev))->smt_idx; @@ -2119,7 +2119,6 @@ static int import_ep(struct c4iw_ep *ep, int iptype, __u8 *peer_ip, ep->rss_qid = cdev->rdev.lldi.rxq_ids[ cxgb4_port_idx(pdev) * step]; set_tcp_window(ep, (struct port_info *)netdev_priv(pdev)); - dev_put(pdev); } else { pdev = get_real_dev(n->dev); ep->l2t = cxgb4_l2t_get(cdev->rdev.lldi.l2t,
ip_dev_find() always returns real net_device address, whether traffic is running on a vlan or real device, if traffic is over vlan, filling endpoint struture with real ndev and an attempt to send a connect request will results in RDMA_CM_EVENT_UNREACHABLE error. This patch fixes the issue by using vlan_dev_real_dev(). Fixes: 830662f6f032 ("RDMA/cxgb4: Add support for active and passive open connection with IPv6 address") Signed-off-by: Anumula Murali Mohan Reddy <anumula@chelsio.com> Signed-off-by: Potnuri Bharat Teja <bharat@chelsio.com> --- drivers/infiniband/hw/cxgb4/cm.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-)