diff options
author | Jakub Kicinski <kuba@kernel.org> | 2022-05-23 21:19:17 -0700 |
---|---|---|
committer | Jakub Kicinski <kuba@kernel.org> | 2022-05-23 21:19:17 -0700 |
commit | 677fb7525331375ba2f90f4bc94a80b9b6e697a3 (patch) | |
tree | d7b9574e487930977c240c5503a3103e32e41cec | |
parent | 1ef0736c0711e2633a59b540931406de626f2836 (diff) | |
parent | 7fb0269720d7d5359bc8349eef908a1c96866b65 (diff) |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
drivers/net/ethernet/cadence/macb_main.c
5cebb40bc955 ("net: macb: Fix PTP one step sync support")
138badbc21a0 ("net: macb: use NAPI for TX completion path")
https://lore.kernel.org/all/20220523111021.31489367@canb.auug.org.au/
net/smc/af_smc.c
75c1edf23b95 ("net/smc: postpone sk_refcnt increment in connect()")
3aba103006bc ("net/smc: align the connect behaviour with TCP")
https://lore.kernel.org/all/20220524114408.4bf1af38@canb.auug.org.au/
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
24 files changed, 192 insertions, 91 deletions
diff --git a/Documentation/admin-guide/sysctl/net.rst b/Documentation/admin-guide/sysctl/net.rst index fa4dcdb283cf..fcd650bdbc7e 100644 --- a/Documentation/admin-guide/sysctl/net.rst +++ b/Documentation/admin-guide/sysctl/net.rst @@ -382,6 +382,15 @@ option is set to SOCK_TXREHASH_DEFAULT (i. e. not overridden by setsockopt). If set to 1 (default), hash rethink is performed on listening socket. If set to 0, hash rethink is not performed. +gro_normal_batch +---------------- + +Maximum number of the segments to batch up on output of GRO. When a packet +exits GRO, either as a coalesced superframe or as an original packet which +GRO has decided not to coalesce, it is placed on a per-NAPI list. This +list is then passed to the stack when the number of segments reaches the +gro_normal_batch limit. + 2. /proc/sys/net/unix - Parameters for Unix domain sockets ---------------------------------------------------------- diff --git a/drivers/net/amt.c b/drivers/net/amt.c index 10455c9b9da0..de4ea518c793 100644 --- a/drivers/net/amt.c +++ b/drivers/net/amt.c @@ -943,7 +943,7 @@ static void amt_req_work(struct work_struct *work) if (amt->status < AMT_STATUS_RECEIVED_ADVERTISEMENT) goto out; - if (amt->req_cnt++ > AMT_MAX_REQ_COUNT) { + if (amt->req_cnt > AMT_MAX_REQ_COUNT) { netdev_dbg(amt->dev, "Gateway is not ready"); amt->qi = AMT_INIT_REQ_TIMEOUT; amt->ready4 = false; @@ -951,13 +951,15 @@ static void amt_req_work(struct work_struct *work) amt->remote_ip = 0; __amt_update_gw_status(amt, AMT_STATUS_INIT, false); amt->req_cnt = 0; + goto out; } spin_unlock_bh(&amt->lock); amt_send_request(amt, false); amt_send_request(amt, true); - amt_update_gw_status(amt, AMT_STATUS_SENT_REQUEST, true); spin_lock_bh(&amt->lock); + __amt_update_gw_status(amt, AMT_STATUS_SENT_REQUEST, true); + amt->req_cnt++; out: exp = min_t(u32, (1 * (1 << amt->req_cnt)), AMT_MAX_REQ_TIMEOUT); mod_delayed_work(amt_wq, &amt->req_wq, msecs_to_jiffies(exp * 1000)); @@ -2696,9 +2698,8 @@ static int amt_rcv(struct sock *sk, struct sk_buff *skb) err = true; goto drop; } - if (amt_advertisement_handler(amt, skb)) - amt->dev->stats.rx_dropped++; - goto out; + err = amt_advertisement_handler(amt, skb); + break; case AMT_MSG_MULTICAST_DATA: if (iph->saddr != amt->remote_ip) { netdev_dbg(amt->dev, "Invalid Relay IP\n"); diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index f245e439f44c..3b7baaeae82c 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -5591,16 +5591,23 @@ static int bond_ethtool_get_ts_info(struct net_device *bond_dev, const struct ethtool_ops *ops; struct net_device *real_dev; struct phy_device *phydev; + int ret = 0; + rcu_read_lock(); real_dev = bond_option_active_slave_get_rcu(bond); + dev_hold(real_dev); + rcu_read_unlock(); + if (real_dev) { ops = real_dev->ethtool_ops; phydev = real_dev->phydev; if (phy_has_tsinfo(phydev)) { - return phy_ts_info(phydev, info); + ret = phy_ts_info(phydev, info); + goto out; } else if (ops->get_ts_info) { - return ops->get_ts_info(real_dev, info); + ret = ops->get_ts_info(real_dev, info); + goto out; } } @@ -5608,7 +5615,9 @@ static int bond_ethtool_get_ts_info(struct net_device *bond_dev, SOF_TIMESTAMPING_SOFTWARE; info->phc_index = -1; - return 0; +out: + dev_put(real_dev); + return ret; } static const struct ethtool_ops bond_ethtool_ops = { diff --git a/drivers/net/dsa/Kconfig b/drivers/net/dsa/Kconfig index 37a3dabdce31..6d1fcb08bba1 100644 --- a/drivers/net/dsa/Kconfig +++ b/drivers/net/dsa/Kconfig @@ -72,7 +72,6 @@ source "drivers/net/dsa/realtek/Kconfig" config NET_DSA_SMSC_LAN9303 tristate - depends on VLAN_8021Q || VLAN_8021Q=n select NET_DSA_TAG_LAN9303 select REGMAP help @@ -82,6 +81,7 @@ config NET_DSA_SMSC_LAN9303 config NET_DSA_SMSC_LAN9303_I2C tristate "SMSC/Microchip LAN9303 3-ports 10/100 ethernet switch in I2C managed mode" depends on I2C + depends on VLAN_8021Q || VLAN_8021Q=n select NET_DSA_SMSC_LAN9303 select REGMAP_I2C help @@ -91,6 +91,7 @@ config NET_DSA_SMSC_LAN9303_I2C config NET_DSA_SMSC_LAN9303_MDIO tristate "SMSC/Microchip LAN9303 3-ports 10/100 ethernet switch in MDIO managed mode" select NET_DSA_SMSC_LAN9303 + depends on VLAN_8021Q || VLAN_8021Q=n help Enable access functions if the SMSC/Microchip LAN9303 is configured for MDIO managed mode. diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index d6cdb97bfb38..d89098f4ede8 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -36,6 +36,7 @@ #include <linux/iopoll.h> #include <linux/phy/phy.h> #include <linux/pm_runtime.h> +#include <linux/ptp_classify.h> #include <linux/reset.h> #include "macb.h" @@ -1122,6 +1123,36 @@ static void macb_tx_error_task(struct work_struct *work) napi_enable(&queue->napi_tx); } +static bool ptp_one_step_sync(struct sk_buff *skb) +{ + struct ptp_header *hdr; + unsigned int ptp_class; + u8 msgtype; + + /* No need to parse packet if PTP TS is not involved */ + if (likely(!(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP))) + goto not_oss; + + /* Identify and return whether PTP one step sync is being processed */ + ptp_class = ptp_classify_raw(skb); + if (ptp_class == PTP_CLASS_NONE) + goto not_oss; + + hdr = ptp_parse_header(skb, ptp_class); + if (!hdr) + goto not_oss; + + if (hdr->flag_field[0] & PTP_FLAG_TWOSTEP) + goto not_oss; + + msgtype = ptp_get_msgtype(hdr, ptp_class); + if (msgtype == PTP_MSGTYPE_SYNC) + return true; + +not_oss: + return false; +} + static int macb_tx_complete(struct macb_queue *queue, int budget) { struct macb *bp = queue->bp; @@ -1158,8 +1189,8 @@ static int macb_tx_complete(struct macb_queue *queue, int budget) /* First, update TX stats if needed */ if (skb) { - if (unlikely(skb_shinfo(skb)->tx_flags & - SKBTX_HW_TSTAMP) && + if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) && + !ptp_one_step_sync(skb) && gem_ptp_do_txstamp(queue, skb, desc) == 0) { /* skb now belongs to timestamp buffer * and will be removed later @@ -2063,7 +2094,8 @@ static unsigned int macb_tx_map(struct macb *bp, ctrl |= MACB_BF(TX_LSO, lso_ctrl); ctrl |= MACB_BF(TX_TCP_SEQ_SRC, seq_ctrl); if ((bp->dev->features & NETIF_F_HW_CSUM) && - skb->ip_summed != CHECKSUM_PARTIAL && !lso_ctrl) + skb->ip_summed != CHECKSUM_PARTIAL && !lso_ctrl && + !ptp_one_step_sync(skb)) ctrl |= MACB_BIT(TX_NOCRC); } else /* Only set MSS/MFS on payload descriptors @@ -2161,7 +2193,7 @@ static int macb_pad_and_fcs(struct sk_buff **skb, struct net_device *ndev) if (!(ndev->features & NETIF_F_HW_CSUM) || !((*skb)->ip_summed != CHECKSUM_PARTIAL) || - skb_shinfo(*skb)->gso_size) /* Not available for GSO */ + skb_shinfo(*skb)->gso_size || ptp_one_step_sync(*skb)) return 0; if (padlen <= 0) { diff --git a/drivers/net/ethernet/cadence/macb_ptp.c b/drivers/net/ethernet/cadence/macb_ptp.c index fb6b27f46b15..9559c16078f9 100644 --- a/drivers/net/ethernet/cadence/macb_ptp.c +++ b/drivers/net/ethernet/cadence/macb_ptp.c @@ -470,8 +470,10 @@ int gem_set_hwtst(struct net_device *dev, struct ifreq *ifr, int cmd) case HWTSTAMP_TX_ONESTEP_SYNC: if (gem_ptp_set_one_step_sync(bp, 1) != 0) return -ERANGE; - fallthrough; + tx_bd_control = TSTAMP_ALL_FRAMES; + break; case HWTSTAMP_TX_ON: + gem_ptp_set_one_step_sync(bp, 0); tx_bd_control = TSTAMP_ALL_FRAMES; break; default: diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c index 4b047255d928..cd9ec80522e7 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c @@ -1097,6 +1097,7 @@ static void dpaa2_eth_free_tx_fd(struct dpaa2_eth_priv *priv, u32 fd_len = dpaa2_fd_get_len(fd); struct dpaa2_sg_entry *sgt; int should_free_skb = 1; + void *tso_hdr; int i; fd_addr = dpaa2_fd_get_addr(fd); @@ -1135,20 +1136,21 @@ static void dpaa2_eth_free_tx_fd(struct dpaa2_eth_priv *priv, sgt = (struct dpaa2_sg_entry *)(buffer_start + priv->tx_data_offset); + /* Unmap the SGT buffer */ + dma_unmap_single(dev, fd_addr, swa->tso.sgt_size, + DMA_BIDIRECTIONAL); + /* Unmap and free the header */ + tso_hdr = dpaa2_iova_to_virt(priv->iommu_domain, dpaa2_sg_get_addr(sgt)); dma_unmap_single(dev, dpaa2_sg_get_addr(sgt), TSO_HEADER_SIZE, DMA_TO_DEVICE); - kfree(dpaa2_iova_to_virt(priv->iommu_domain, dpaa2_sg_get_addr(sgt))); + kfree(tso_hdr); /* Unmap the other SG entries for the data */ for (i = 1; i < swa->tso.num_sg; i++) dma_unmap_single(dev, dpaa2_sg_get_addr(&sgt[i]), dpaa2_sg_get_len(&sgt[i]), DMA_TO_DEVICE); - /* Unmap the SGT buffer */ - dma_unmap_single(dev, fd_addr, swa->sg.sgt_size, - DMA_BIDIRECTIONAL); - if (!swa->tso.is_last_fd) should_free_skb = 0; } else { diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 9d7ef993ce45..a90275143d87 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -3876,9 +3876,11 @@ fec_probe(struct platform_device *pdev) mutex_init(&fep->ptp_clk_mutex); /* clk_ref is optional, depends on board */ - fep->clk_ref = devm_clk_get(&pdev->dev, "enet_clk_ref"); - if (IS_ERR(fep->clk_ref)) - fep->clk_ref = NULL; + fep->clk_ref = devm_clk_get_optional(&pdev->dev, "enet_clk_ref"); + if (IS_ERR(fep->clk_ref)) { + ret = PTR_ERR(fep->clk_ref); + goto failed_clk; + } fep->clk_ref_rate = clk_get_rate(fep->clk_ref); /* clk_2x_txclk is optional, depends on board */ diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.c index f7dc7d825f63..4daf6bf291ec 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.c @@ -386,7 +386,7 @@ static int alloc_wqes_shadow(struct hinic_wq *wq) return -ENOMEM; wq->shadow_idx = devm_kcalloc(&pdev->dev, wq->num_q_pages, - sizeof(wq->prod_idx), GFP_KERNEL); + sizeof(*wq->shadow_idx), GFP_KERNEL); if (!wq->shadow_idx) goto err_shadow_idx; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c index 9f1759593b94..2fc51dc5eb0b 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c @@ -1084,8 +1084,9 @@ static int stmmac_test_rxp(struct stmmac_priv *priv) unsigned char addr[ETH_ALEN] = {0xde, 0xad, 0xbe, 0xef, 0x00, 0x00}; struct tc_cls_u32_offload cls_u32 = { }; struct stmmac_packet_attrs attr = { }; - struct tc_action **actions, *act; + struct tc_action **actions; struct tc_u32_sel *sel; + struct tcf_gact *gact; struct tcf_exts *exts; int ret, i, nk = 1; @@ -1110,8 +1111,8 @@ static int stmmac_test_rxp(struct stmmac_priv *priv) goto cleanup_exts; } - act = kcalloc(nk, sizeof(*act), GFP_KERNEL); - if (!act) { + gact = kcalloc(nk, sizeof(*gact), GFP_KERNEL); + if (!gact) { ret = -ENOMEM; goto cleanup_actions; } @@ -1126,9 +1127,7 @@ static int stmmac_test_rxp(struct stmmac_priv *priv) exts->nr_actions = nk; exts->actions = actions; for (i = 0; i < nk; i++) { - struct tcf_gact *gact = to_gact(&act[i]); - - actions[i] = &act[i]; + actions[i] = (struct tc_action *)&gact[i]; gact->tcf_action = TC_ACT_SHOT; } @@ -1152,7 +1151,7 @@ static int stmmac_test_rxp(struct stmmac_priv *priv) stmmac_tc_setup_cls_u32(priv, priv, &cls_u32); cleanup_act: - kfree(act); + kfree(gact); cleanup_actions: kfree(actions); cleanup_exts: diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index 27f6bbca6619..15ebd5426604 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -2643,7 +2643,10 @@ static int netvsc_suspend(struct hv_device *dev) /* Save the current config info */ ndev_ctx->saved_netvsc_dev_info = netvsc_devinfo_get(nvdev); - + if (!ndev_ctx->saved_netvsc_dev_info) { + ret = -ENOMEM; + goto out; + } ret = netvsc_detach(net, nvdev); out: rtnl_unlock(); diff --git a/drivers/nfc/st21nfca/se.c b/drivers/nfc/st21nfca/se.c index c922f10d0d7b..7e213f8ddc98 100644 --- a/drivers/nfc/st21nfca/se.c +++ b/drivers/nfc/st21nfca/se.c @@ -241,7 +241,7 @@ int st21nfca_hci_se_io(struct nfc_hci_dev *hdev, u32 se_idx, } EXPORT_SYMBOL(st21nfca_hci_se_io); -static void st21nfca_se_wt_timeout(struct timer_list *t) +static void st21nfca_se_wt_work(struct work_struct *work) { /* * No answer from the secure element @@ -254,8 +254,9 @@ static void st21nfca_se_wt_timeout(struct timer_list *t) */ /* hardware reset managed through VCC_UICC_OUT power supply */ u8 param = 0x01; - struct st21nfca_hci_info *info = from_timer(info, t, - se_info.bwi_timer); + struct st21nfca_hci_info *info = container_of(work, + struct st21nfca_hci_info, + se_info.timeout_work); info->se_info.bwi_active = false; @@ -271,6 +272,13 @@ static void st21nfca_se_wt_timeout(struct timer_list *t) info->se_info.cb(info->se_info.cb_context, NULL, 0, -ETIME); } +static void st21nfca_se_wt_timeout(struct timer_list *t) +{ + struct st21nfca_hci_info *info = from_timer(info, t, se_info.bwi_timer); + + schedule_work(&info->se_info.timeout_work); +} + static void st21nfca_se_activation_timeout(struct timer_list *t) { struct st21nfca_hci_info *info = from_timer(info, t, @@ -360,6 +368,7 @@ int st21nfca_apdu_reader_event_received(struct nfc_hci_dev *hdev, switch (event) { case ST21NFCA_EVT_TRANSMIT_DATA: del_timer_sync(&info->se_info.bwi_timer); + cancel_work_sync(&info->se_info.timeout_work); info->se_info.bwi_active = false; r = nfc_hci_send_event(hdev, ST21NFCA_DEVICE_MGNT_GATE, ST21NFCA_EVT_SE_END_OF_APDU_TRANSFER, NULL, 0); @@ -389,6 +398,7 @@ void st21nfca_se_init(struct nfc_hci_dev *hdev) struct st21nfca_hci_info *info = nfc_hci_get_clientdata(hdev); init_completion(&info->se_info.req_completion); + INIT_WORK(&info->se_info.timeout_work, st21nfca_se_wt_work); /* initialize timers */ timer_setup(&info->se_info.bwi_timer, st21nfca_se_wt_timeout, 0); info->se_info.bwi_active = false; @@ -416,6 +426,7 @@ void st21nfca_se_deinit(struct nfc_hci_dev *hdev) if (info->se_info.se_active) del_timer_sync(&info->se_info.se_active_timer); + cancel_work_sync(&info->se_info.timeout_work); info->se_info.bwi_active = false; info->se_info.se_active = false; } diff --git a/drivers/nfc/st21nfca/st21nfca.h b/drivers/nfc/st21nfca/st21nfca.h index cb6ad916be91..ae6771cc9894 100644 --- a/drivers/nfc/st21nfca/st21nfca.h +++ b/drivers/nfc/st21nfca/st21nfca.h @@ -141,6 +141,7 @@ struct st21nfca_se_info { se_io_cb_t cb; void *cb_context; + struct work_struct timeout_work; }; struct st21nfca_hci_info { diff --git a/include/linux/ptp_classify.h b/include/linux/ptp_classify.h index fefa7790dc46..2b6ea36ad162 100644 --- a/include/linux/ptp_classify.h +++ b/include/linux/ptp_classify.h @@ -43,6 +43,9 @@ #define OFF_PTP_SOURCE_UUID 22 /* PTPv1 only */ #define OFF_PTP_SEQUENCE_ID 30 +/* PTP header flag fields */ +#define PTP_FLAG_TWOSTEP BIT(1) + /* Below defines should actually be removed at some point in time. */ #define IP6_HLEN 40 #define UDP_HLEN 8 diff --git a/include/trace/events/rxrpc.h b/include/trace/events/rxrpc.h index 66915b872a44..d20bf4aa0204 100644 --- a/include/trace/events/rxrpc.h +++ b/include/trace/events/rxrpc.h @@ -1330,7 +1330,7 @@ TRACE_EVENT(rxrpc_call_reset, __entry->call_serial = call->rx_serial; __entry->conn_serial = call->conn->hi_serial; __entry->tx_seq = call->tx_hard_ack; - __entry->rx_seq = call->ackr_seen; + __entry->rx_seq = call->rx_hard_ack; ), TP_printk("c=%08x %08x:%08x r=%08x/%08x tx=%08x rx=%08x", diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 882a7df13005..ac06c9724c7f 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -943,10 +943,11 @@ static void create_le_conn_complete(struct hci_dev *hdev, void *data, int err) bt_dev_err(hdev, "request failed to create LE connection: err %d", err); - if (!conn) + /* Check if connection is still pending */ + if (conn != hci_lookup_le_connect(hdev)) goto done; - hci_le_conn_failed(conn, err); + hci_conn_failed(conn, err); done: hci_dev_unlock(hdev); diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 0270e597c285..af17dfb20e01 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -5632,10 +5632,12 @@ static void le_conn_complete_evt(struct hci_dev *hdev, u8 status, status = HCI_ERROR_INVALID_PARAMETERS; } - if (status) { - hci_conn_failed(conn, status); + /* All connection failure handling is taken care of by the + * hci_conn_failed function which is triggered by the HCI + * request completion callbacks used for connecting. + */ + if (status) goto unlock; - } if (conn->dst_type == ADDR_LE_DEV_PUBLIC) addr_type = BDADDR_LE_PUBLIC; diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index dcc0ec0bf3de..571436064cd6 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -668,13 +668,12 @@ struct rxrpc_call { spinlock_t input_lock; /* Lock for packet input to this call */ - /* receive-phase ACK management */ + /* Receive-phase ACK management (ACKs we send). */ u8 ackr_reason; /* reason to ACK */ rxrpc_serial_t ackr_serial; /* serial of packet being ACK'd */ - rxrpc_serial_t ackr_first_seq; /* first sequence number received */ - rxrpc_seq_t ackr_prev_seq; /* previous sequence number received */ - rxrpc_seq_t ackr_consumed; /* Highest packet shown consumed */ - rxrpc_seq_t ackr_seen; /* Highest packet shown seen */ + rxrpc_seq_t ackr_highest_seq; /* Higest sequence number received */ + atomic_t ackr_nr_unacked; /* Number of unacked packets */ + atomic_t ackr_nr_consumed; /* Number of packets needing hard ACK */ /* RTT management */ rxrpc_serial_t rtt_serial[4]; /* Serial number of DATA or PING sent */ @@ -684,8 +683,10 @@ struct rxrpc_call { #define RXRPC_CALL_RTT_AVAIL_MASK 0xf #define RXRPC_CALL_RTT_PEND_SHIFT 8 - /* transmission-phase ACK management */ + /* Transmission-phase ACK management (ACKs we've received). */ ktime_t acks_latest_ts; /* Timestamp of latest ACK received */ + rxrpc_seq_t acks_first_seq; /* first sequence number received */ + rxrpc_seq_t acks_prev_seq; /* Highest previousPacket received */ rxrpc_seq_t acks_lowest_nak; /* Lowest NACK in the buffer (or ==tx_hard_ack) */ rxrpc_seq_t acks_lost_top; /* tx_top at the time lost-ack ping sent */ rxrpc_serial_t acks_lost_ping; /* Serial number of probe ACK */ diff --git a/net/rxrpc/call_event.c b/net/rxrpc/call_event.c index e426f6831aab..f8ecad2b730e 100644 --- a/net/rxrpc/call_event.c +++ b/net/rxrpc/call_event.c @@ -406,7 +406,8 @@ recheck_state: goto recheck_state; } - if (test_and_clear_bit(RXRPC_CALL_EV_RESEND, &call->events)) { + if (test_and_clear_bit(RXRPC_CALL_EV_RESEND, &call->events) && + call->state != RXRPC_CALL_CLIENT_RECV_REPLY) { rxrpc_resend(call, now); goto recheck_state; } diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index 16c0af41c202..721d847ba92b 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -412,8 +412,8 @@ static void rxrpc_input_data(struct rxrpc_call *call, struct sk_buff *skb) { struct rxrpc_skb_priv *sp = rxrpc_skb(skb); enum rxrpc_call_state state; - unsigned int j, nr_subpackets; - rxrpc_serial_t serial = sp->hdr.serial, ack_serial = 0; + unsigned int j, nr_subpackets, nr_unacked = 0; + rxrpc_serial_t serial = sp->hdr.serial, ack_serial = serial; rxrpc_seq_t seq0 = sp->hdr.seq, hard_ack; bool immediate_ack = false, jumbo_bad = false; u8 ack = 0; @@ -453,7 +453,6 @@ static void rxrpc_input_data(struct rxrpc_call *call, struct sk_buff *skb) !rxrpc_receiving_reply(call)) goto unlock; - call->ackr_prev_seq = seq0; hard_ack = READ_ONCE(call->rx_hard_ack); nr_subpackets = sp->nr_subpackets; @@ -534,6 +533,9 @@ static void rxrpc_input_data(struct rxrpc_call *call, struct sk_buff *skb) ack_serial = serial; } + if (after(seq0, call->ackr_highest_seq)) + call->ackr_highest_seq = seq0; + /* Queue the packet. We use a couple of memory barriers here as need * to make sure that rx_top is perceived to be set after the buffer * pointer and that the buffer pointer is set after the annotation and @@ -567,6 +569,8 @@ static void rxrpc_input_data(struct rxrpc_call *call, struct sk_buff *skb) sp = NULL; } + nr_unacked++; + if (last) { set_bit(RXRPC_CALL_RX_LAST, &call->flags); if (!ack) { @@ -586,9 +590,14 @@ static void rxrpc_input_data(struct rxrpc_call *call, struct sk_buff *skb) } call->rx_expect_next = seq + 1; } + if (!ack) + ack_serial = serial; } ack: + if (atomic_add_return(nr_unacked, &call->ackr_nr_unacked) > 2 && !ack) + ack = RXRPC_ACK_IDLE; + if (ack) rxrpc_propose_ACK(call, ack, ack_serial, immediate_ack, true, @@ -812,7 +821,7 @@ static void rxrpc_input_soft_acks(struct rxrpc_call *call, u8 *acks, static bool rxrpc_is_ack_valid(struct rxrpc_call *call, rxrpc_seq_t first_pkt, rxrpc_seq_t prev_pkt) { - rxrpc_seq_t base = READ_ONCE(call->ackr_first_seq); + rxrpc_seq_t base = READ_ONCE(call->acks_first_seq); if (after(first_pkt, base)) return true; /* The window advanced */ @@ -820,7 +829,7 @@ static bool rxrpc_is_ack_valid(struct rxrpc_call *call, if (before(first_pkt, base)) return false; /* firstPacket regressed */ - if (after_eq(prev_pkt, call->ackr_prev_seq)) + if (after_eq(prev_pkt, call->acks_prev_seq)) return true; /* previousPacket hasn't regressed. */ /* Some rx implementations put a serial number in previousPacket. */ @@ -933,8 +942,8 @@ static void rxrpc_input_ack(struct rxrpc_call *call, struct sk_buff *skb) /* Discard any out-of-order or duplicate ACKs (outside lock). */ if (!rxrpc_is_ack_valid(call, first_soft_ack, prev_pkt)) { trace_rxrpc_rx_discard_ack(call->debug_id, ack_serial, - first_soft_ack, call->ackr_first_seq, - prev_pkt, call->ackr_prev_seq); + first_soft_ack, call->acks_first_seq, + prev_pkt, call->acks_prev_seq); return; } @@ -949,14 +958,14 @@ static void rxrpc_input_ack(struct rxrpc_call *call, struct sk_buff *skb) /* Discard any out-of-order or duplicate ACKs (inside lock). */ if (!rxrpc_is_ack_valid(call, first_soft_ack, prev_pkt)) { trace_rxrpc_rx_discard_ack(call->debug_id, ack_serial, - first_soft_ack, call->ackr_first_seq, - prev_pkt, call->ackr_prev_seq); + first_soft_ack, call->acks_first_seq, + prev_pkt, call->acks_prev_seq); goto out; } call->acks_latest_ts = skb->tstamp; - call->ackr_first_seq = first_soft_ack; - call->ackr_prev_seq = prev_pkt; + call->acks_first_seq = first_soft_ack; + call->acks_prev_seq = prev_pkt; /* Parse rwind and mtu sizes if provided. */ if (buf.info.rxMTU) diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c index a45c83f22236..9683617db704 100644 --- a/net/rxrpc/output.c +++ b/net/rxrpc/output.c @@ -74,11 +74,18 @@ static size_t rxrpc_fill_out_ack(struct rxrpc_connection *conn, u8 reason) { rxrpc_serial_t serial; + unsigned int tmp; rxrpc_seq_t hard_ack, top, seq; int ix; u32 mtu, jmax; u8 *ackp = pkt->acks; + tmp = atomic_xchg(&call->ackr_nr_unacked, 0); + tmp |= atomic_xchg(&call->ackr_nr_consumed, 0); + if (!tmp && (reason == RXRPC_ACK_DELAY || + reason == RXRPC_ACK_IDLE)) + return 0; + /* Barrier against rxrpc_input_data(). */ serial = call->ackr_serial; hard_ack = READ_ONCE(call->rx_hard_ack); @@ -89,7 +96,7 @@ static size_t rxrpc_fill_out_ack(struct rxrpc_connection *conn, pkt->ack.bufferSpace = htons(8); pkt->ack.maxSkew = htons(0); pkt->ack.firstPacket = htonl(hard_ack + 1); - pkt->ack.previousPacket = htonl(call->ackr_prev_seq); + pkt->ack.previousPacket = htonl(call->ackr_highest_seq); pkt->ack.serial = htonl(serial); pkt->ack.reason = reason; pkt->ack.nAcks = top - hard_ack; @@ -223,6 +230,10 @@ int rxrpc_send_ack_packet(struct rxrpc_call *call, bool ping, n = rxrpc_fill_out_ack(conn, call, pkt, &hard_ack, &top, reason); spin_unlock_bh(&call->lock); + if (n == 0) { + kfree(pkt); + return 0; + } iov[0].iov_base = pkt; iov[0].iov_len = sizeof(pkt->whdr) + sizeof(pkt->ack) + n; @@ -259,13 +270,6 @@ int rxrpc_send_ack_packet(struct rxrpc_call *call, bool ping, ntohl(pkt->ack.serial), false, true, rxrpc_propose_ack_retry_tx); - } else { - spin_lock_bh(&call->lock); - if (after(hard_ack, call->ackr_consumed)) - call->ackr_consumed = hard_ack; - if (after(top, call->ackr_seen)) - call->ackr_seen = top; - spin_unlock_bh(&call->lock); } rxrpc_set_keepalive(call); diff --git a/net/rxrpc/recvmsg.c b/net/rxrpc/recvmsg.c index eca6dda26c77..250f23bc1c07 100644 --- a/net/rxrpc/recvmsg.c +++ b/net/rxrpc/recvmsg.c @@ -260,11 +260,9 @@ static void rxrpc_rotate_rx_window(struct rxrpc_call *call) rxrpc_end_rx_phase(call, serial); } else { /* Check to see if there's an ACK that needs sending. */ - if (after_eq(hard_ack, call->ackr_consumed + 2) || - after_eq(top, call->ackr_seen + 2) || - (hard_ack == top && after(hard_ack, call->ackr_consumed))) - rxrpc_propose_ACK(call, RXRPC_ACK_DELAY, serial, - true, true, + if (atomic_inc_return(&call->ackr_nr_consumed) > 2) + rxrpc_propose_ACK(call, RXRPC_ACK_IDLE, serial, + true, false, rxrpc_propose_ack_rotate_rx); if (call->ackr_reason && call->ackr_reason != RXRPC_ACK_DELAY) rxrpc_send_ack_packet(call, false, NULL); diff --git a/net/rxrpc/sysctl.c b/net/rxrpc/sysctl.c index 540351d6a5f4..555e0910786b 100644 --- a/net/rxrpc/sysctl.c +++ b/net/rxrpc/sysctl.c @@ -12,7 +12,7 @@ static struct ctl_table_header *rxrpc_sysctl_reg_table; static const unsigned int four = 4; -static const unsigned int thirtytwo = 32; +static const unsigned int max_backlog = RXRPC_BACKLOG_MAX - 1; static const unsigned int n_65535 = 65535; static const unsigned int n_max_acks = RXRPC_RXTX_BUFF_SIZE - 1; static const unsigned long one_jiffy = 1; @@ -89,7 +89,7 @@ static struct ctl_table rxrpc_sysctl_table[] = { .mode = 0644, .proc_handler = proc_dointvec_minmax, .extra1 = (void *)&four, - .extra2 = (void *)&thirtytwo, + .extra2 = (void *)&max_backlog, }, { .procname = "rx_window_size", diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 5f70642a8044..1a556f47f89e 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -1584,11 +1584,11 @@ static int smc_connect(struct socket *sock, struct sockaddr *addr, if (rc && rc != -EINPROGRESS) goto out; - sock_hold(&smc->sk); /* sock put in passive closing */ if (smc->use_fallback) { sock->state = rc ? SS_CONNECTING : SS_CONNECTED; goto out; } + sock_hold(&smc->sk); /* sock put in passive closing */ if (flags & O_NONBLOCK) { if (queue_work(smc_hs_wq, &smc->connect_work)) smc->connect_nonblock = 1; @@ -2118,13 +2118,13 @@ static int smc_listen_rdma_reg(struct smc_sock *new_smc, bool local_first) return 0; } -static void smc_find_rdma_v2_device_serv(struct smc_sock *new_smc, - struct smc_clc_msg_proposal *pclc, - struct smc_init_info *ini) +static int smc_find_rdma_v2_device_serv(struct smc_sock *new_smc, + struct smc_clc_msg_proposal *pclc, + struct smc_init_info *ini) { struct smc_clc_v2_extension *smc_v2_ext; u8 smcr_version; - int rc; + int rc = 0; if (!(ini->smcr_version & SMC_V2) || !smcr_indicated(ini->smc_type_v2)) goto not_found; @@ -2142,26 +2142,31 @@ static void smc_find_rdma_v2_device_serv(struct smc_sock *new_smc, ini->smcrv2.saddr = new_smc->clcsock->sk->sk_rcv_saddr; ini->smcrv2.daddr = smc_ib_gid_to_ipv4(smc_v2_ext->roce); rc = smc_find_rdma_device(new_smc, ini); - if (rc) { - smc_find_ism_store_rc(rc, ini); + if (rc) goto not_found; - } + if (!ini->smcrv2.uses_gateway) memcpy(ini->smcrv2.nexthop_mac, pclc->lcl.mac, ETH_ALEN); smcr_version = ini->smcr_version; ini->smcr_version = SMC_V2; rc = smc_listen_rdma_init(new_smc, ini); - if (!rc) - rc = smc_listen_rdma_reg(new_smc, ini->first_contact_local); - if (!rc) - return; - ini->smcr_version = smcr_version; - smc_find_ism_store_rc(rc, ini); + if (rc) { + ini->smcr_version = smcr_version; + goto not_found; + } + rc = smc_listen_rdma_reg(new_smc, ini->first_contact_local); + if (rc) { + ini->smcr_version = smcr_version; + goto not_found; + } + return 0; not_found: + rc = rc ?: SMC_CLC_DECL_NOSMCDEV; ini->smcr_version &= ~SMC_V2; ini->check_smcrv2 = false; + return rc; } static int smc_find_rdma_v1_device_serv(struct smc_sock *new_smc, @@ -2194,6 +2199,7 @@ static int smc_listen_find_device(struct smc_sock *new_smc, struct smc_init_info *ini) { int prfx_rc; + int rc; /* check for ISM device matching V2 proposed device */ smc_find_ism_v2_device_serv(new_smc, pclc, ini); @@ -2221,14 +2227,18 @@ static int smc_listen_find_device(struct smc_sock *new_smc, return ini->rc ?: SMC_CLC_DECL_NOSMCDDEV; /* check if RDMA V2 is available */ - smc_find_rdma_v2_device_serv(new_smc, pclc, ini); - if (ini->smcrv2.ib_dev_v2) + rc = smc_find_rdma_v2_device_serv(new_smc, pclc, ini); + if (!rc) return 0; + /* skip V1 check if V2 is unavailable for non-Device reason */ + if (rc != SMC_CLC_DECL_NOSMCDEV && + rc != SMC_CLC_DECL_NOSMCRDEV && + rc != SMC_CLC_DECL_NOSMCDDEV) + return rc; + /* check if RDMA V1 is available */ if (!prfx_rc) { - int rc; - rc = smc_find_rdma_v1_device_serv(new_smc, pclc, ini); smc_find_ism_store_rc(rc, ini); return (!rc) ? 0 : ini->rc; |