diff --git a/CREDITS b/CREDITS index a357f9c..a687c3c 100644 --- a/CREDITS +++ b/CREDITS @@ -3222,6 +3222,10 @@ D: AIC5800 IEEE 1394, RAW I/O on 1394 D: Starter of Linux1394 effort S: ask per mail for current address +N: Boris Pismenny +E: borisp@mellanox.com +D: Kernel TLS implementation and offload support. + N: Nicolas Pitre E: nico@fluxnic.net D: StrongARM SA1100 support integrator & hacker @@ -4168,6 +4172,9 @@ S: 1513 Brewster Dr. S: Carrollton, TX 75010 S: USA +N: Dave Watson +D: Kernel TLS implementation. + N: Tim Waugh E: tim@cyberelk.net D: Co-architect of the parallel-port sharing system diff --git a/MAINTAINERS b/MAINTAINERS index adcbb09..09b34bb 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -937,7 +937,7 @@ S: Maintained F: drivers/gpio/gpio-altera.c ALTERA TRIPLE SPEED ETHERNET DRIVER -M: Joyce Ooi +M: Boon Khai Ng L: netdev@vger.kernel.org S: Maintained F: drivers/net/ethernet/altera/ @@ -17848,7 +17848,6 @@ F: net/ipv6/syncookies.c F: net/ipv6/tcp*.c NETWORKING [TLS] -M: Boris Pismenny M: John Fastabend M: Jakub Kicinski L: netdev@vger.kernel.org @@ -20878,8 +20877,8 @@ S: Maintained F: drivers/firmware/qcom/qcom_qseecom_uefisecapp.c QUALCOMM RMNET DRIVER -M: Subash Abhinov Kasiviswanathan -M: Sean Tranchetti +M: Subash Abhinov Kasiviswanathan +M: Sean Tranchetti L: netdev@vger.kernel.org S: Maintained F: Documentation/networking/device_drivers/cellular/qualcomm/rmnet.rst diff --git a/drivers/atm/atmtcp.c b/drivers/atm/atmtcp.c index eeae160..fa3c76a 100644 --- a/drivers/atm/atmtcp.c +++ b/drivers/atm/atmtcp.c @@ -279,6 +279,19 @@ static struct atm_vcc *find_vcc(struct atm_dev *dev, short vpi, int vci) return NULL; } +static int atmtcp_c_pre_send(struct atm_vcc *vcc, struct sk_buff *skb) +{ + struct atmtcp_hdr *hdr; + + if (skb->len < sizeof(struct atmtcp_hdr)) + return -EINVAL; + + hdr = (struct atmtcp_hdr *)skb->data; + if (hdr->length == ATMTCP_HDR_MAGIC) + return -EINVAL; + + return 0; +} static int atmtcp_c_send(struct atm_vcc *vcc,struct sk_buff *skb) { @@ -288,9 +301,6 @@ static int atmtcp_c_send(struct atm_vcc *vcc,struct sk_buff *skb) struct sk_buff *new_skb; int result = 0; - if (skb->len < sizeof(struct atmtcp_hdr)) - goto done; - dev = vcc->dev_data; hdr = (struct atmtcp_hdr *) skb->data; if (hdr->length == ATMTCP_HDR_MAGIC) { @@ -347,6 +357,7 @@ static const struct atmdev_ops atmtcp_v_dev_ops = { static const struct atmdev_ops atmtcp_c_dev_ops = { .close = atmtcp_c_close, + .pre_send = atmtcp_c_pre_send, .send = atmtcp_c_send }; diff --git a/drivers/isdn/hardware/mISDN/hfcpci.c b/drivers/isdn/hardware/mISDN/hfcpci.c index 2b05722..ea8a0ab 100644 --- a/drivers/isdn/hardware/mISDN/hfcpci.c +++ b/drivers/isdn/hardware/mISDN/hfcpci.c @@ -39,12 +39,13 @@ #include "hfc_pci.h" +static void hfcpci_softirq(struct timer_list *unused); static const char *hfcpci_revision = "2.0"; static int HFC_cnt; static uint debug; static uint poll, tics; -static struct timer_list hfc_tl; +static DEFINE_TIMER(hfc_tl, hfcpci_softirq); static unsigned long hfc_jiffies; MODULE_AUTHOR("Karsten Keil"); @@ -2305,8 +2306,7 @@ hfcpci_softirq(struct timer_list *unused) hfc_jiffies = jiffies + 1; else hfc_jiffies += tics; - hfc_tl.expires = hfc_jiffies; - add_timer(&hfc_tl); + mod_timer(&hfc_tl, hfc_jiffies); } static int __init @@ -2332,10 +2332,8 @@ HFC_init(void) if (poll != HFCPCI_BTRANS_THRESHOLD) { printk(KERN_INFO "%s: Using alternative poll value of %d\n", __func__, poll); - timer_setup(&hfc_tl, hfcpci_softirq, 0); - hfc_tl.expires = jiffies + tics; - hfc_jiffies = hfc_tl.expires; - add_timer(&hfc_tl); + hfc_jiffies = jiffies + tics; + mod_timer(&hfc_tl, hfc_jiffies); } else tics = 0; /* indicate the use of controller's timer */ diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 207a8bb..31e3d82 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -8016,7 +8016,8 @@ static int __bnxt_reserve_rings(struct bnxt *bp) } rx_rings = min_t(int, rx_rings, hwr.grp); hwr.cp = min_t(int, hwr.cp, bp->cp_nr_rings); - if (hwr.stat > bnxt_get_ulp_stat_ctxs(bp)) + if (bnxt_ulp_registered(bp->edev) && + hwr.stat > bnxt_get_ulp_stat_ctxs(bp)) hwr.stat -= bnxt_get_ulp_stat_ctxs(bp); hwr.cp = min_t(int, hwr.cp, hwr.stat); rc = bnxt_trim_rings(bp, &rx_rings, &hwr.tx, hwr.cp, sh); @@ -8024,6 +8025,11 @@ static int __bnxt_reserve_rings(struct bnxt *bp) hwr.rx = rx_rings << 1; tx_cp = bnxt_num_tx_to_cp(bp, hwr.tx); hwr.cp = sh ? max_t(int, tx_cp, rx_rings) : tx_cp + rx_rings; + if (hwr.tx != bp->tx_nr_rings) { + netdev_warn(bp->dev, + "Able to reserve only %d out of %d requested TX rings\n", + hwr.tx, bp->tx_nr_rings); + } bp->tx_nr_rings = hwr.tx; /* If we cannot reserve all the RX rings, reset the RSS map only @@ -12851,6 +12857,17 @@ static int bnxt_set_xps_mapping(struct bnxt *bp) return rc; } +static int bnxt_tx_nr_rings(struct bnxt *bp) +{ + return bp->num_tc ? bp->tx_nr_rings_per_tc * bp->num_tc : + bp->tx_nr_rings_per_tc; +} + +static int bnxt_tx_nr_rings_per_tc(struct bnxt *bp) +{ + return bp->num_tc ? bp->tx_nr_rings / bp->num_tc : bp->tx_nr_rings; +} + static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init) { int rc = 0; @@ -12868,6 +12885,13 @@ static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init) if (rc) return rc; + /* Make adjustments if reserved TX rings are less than requested */ + bp->tx_nr_rings -= bp->tx_nr_rings_xdp; + bp->tx_nr_rings_per_tc = bnxt_tx_nr_rings_per_tc(bp); + if (bp->tx_nr_rings_xdp) { + bp->tx_nr_rings_xdp = bp->tx_nr_rings_per_tc; + bp->tx_nr_rings += bp->tx_nr_rings_xdp; + } rc = bnxt_alloc_mem(bp, irq_re_init); if (rc) { netdev_err(bp->dev, "bnxt_alloc_mem err: %x\n", rc); @@ -16325,7 +16349,7 @@ static void bnxt_trim_dflt_sh_rings(struct bnxt *bp) bp->cp_nr_rings = min_t(int, bp->tx_nr_rings_per_tc, bp->rx_nr_rings); bp->rx_nr_rings = bp->cp_nr_rings; bp->tx_nr_rings_per_tc = bp->cp_nr_rings; - bp->tx_nr_rings = bp->tx_nr_rings_per_tc; + bp->tx_nr_rings = bnxt_tx_nr_rings(bp); } static int bnxt_set_dflt_rings(struct bnxt *bp, bool sh) @@ -16357,7 +16381,7 @@ static int bnxt_set_dflt_rings(struct bnxt *bp, bool sh) bnxt_trim_dflt_sh_rings(bp); else bp->cp_nr_rings = bp->tx_nr_rings_per_tc + bp->rx_nr_rings; - bp->tx_nr_rings = bp->tx_nr_rings_per_tc; + bp->tx_nr_rings = bnxt_tx_nr_rings(bp); avail_msix = bnxt_get_max_func_irqs(bp) - bp->cp_nr_rings; if (avail_msix >= BNXT_MIN_ROCE_CP_RINGS) { @@ -16370,7 +16394,7 @@ static int bnxt_set_dflt_rings(struct bnxt *bp, bool sh) rc = __bnxt_reserve_rings(bp); if (rc && rc != -ENODEV) netdev_warn(bp->dev, "Unable to reserve tx rings\n"); - bp->tx_nr_rings_per_tc = bp->tx_nr_rings; + bp->tx_nr_rings_per_tc = bnxt_tx_nr_rings_per_tc(bp); if (sh) bnxt_trim_dflt_sh_rings(bp); @@ -16379,7 +16403,7 @@ static int bnxt_set_dflt_rings(struct bnxt *bp, bool sh) rc = __bnxt_reserve_rings(bp); if (rc && rc != -ENODEV) netdev_warn(bp->dev, "2nd rings reservation failed.\n"); - bp->tx_nr_rings_per_tc = bp->tx_nr_rings; + bp->tx_nr_rings_per_tc = bnxt_tx_nr_rings_per_tc(bp); } if (BNXT_CHIP_TYPE_NITRO_A0(bp)) { bp->rx_nr_rings++; @@ -16413,7 +16437,7 @@ static int bnxt_init_dflt_ring_mode(struct bnxt *bp) if (rc) goto init_dflt_ring_err; - bp->tx_nr_rings_per_tc = bp->tx_nr_rings; + bp->tx_nr_rings_per_tc = bnxt_tx_nr_rings_per_tc(bp); bnxt_set_dflt_rfs(bp); diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 9693f02..16d28a8 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -3090,7 +3090,7 @@ static void gem_update_stats(struct macb *bp) /* Add GEM_OCTTXH, GEM_OCTRXH */ val = bp->macb_reg_readl(bp, offset + 4); bp->ethtool_stats[i] += ((u64)val) << 32; - *(p++) += ((u64)val) << 32; + *p += ((u64)val) << 32; } } @@ -5399,19 +5399,16 @@ static void macb_remove(struct platform_device *pdev) if (dev) { bp = netdev_priv(dev); + unregister_netdev(dev); phy_exit(bp->sgmii_phy); mdiobus_unregister(bp->mii_bus); mdiobus_free(bp->mii_bus); - unregister_netdev(dev); + device_set_wakeup_enable(&bp->pdev->dev, 0); cancel_work_sync(&bp->hresp_err_bh_work); pm_runtime_disable(&pdev->dev); pm_runtime_dont_use_autosuspend(&pdev->dev); - if (!pm_runtime_suspended(&pdev->dev)) { - macb_clks_disable(bp->pclk, bp->hclk, bp->tx_clk, - bp->rx_clk, bp->tsu_clk); - pm_runtime_set_suspended(&pdev->dev); - } + pm_runtime_set_suspended(&pdev->dev); phylink_destroy(bp->phylink); free_netdev(dev); } diff --git a/drivers/net/ethernet/dlink/dl2k.c b/drivers/net/ethernet/dlink/dl2k.c index cc60ee4..6bbf6e5 100644 --- a/drivers/net/ethernet/dlink/dl2k.c +++ b/drivers/net/ethernet/dlink/dl2k.c @@ -1099,7 +1099,7 @@ get_stats (struct net_device *dev) dev->stats.rx_bytes += dr32(OctetRcvOk); dev->stats.tx_bytes += dr32(OctetXmtOk); - dev->stats.multicast = dr32(McstFramesRcvdOk); + dev->stats.multicast += dr32(McstFramesRcvdOk); dev->stats.collisions += dr32(SingleColFrames) + dr32(MultiColFrames); diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index 2098f00..8a8a01a 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -510,6 +510,7 @@ enum ice_pf_flags { ICE_FLAG_LINK_LENIENT_MODE_ENA, ICE_FLAG_PLUG_AUX_DEV, ICE_FLAG_UNPLUG_AUX_DEV, + ICE_FLAG_AUX_DEV_CREATED, ICE_FLAG_MTU_CHANGED, ICE_FLAG_GNSS, /* GNSS successfully initialized */ ICE_FLAG_DPLL, /* SyncE/PTP dplls initialized */ diff --git a/drivers/net/ethernet/intel/ice/ice_adapter.c b/drivers/net/ethernet/intel/ice/ice_adapter.c index 9e4adc4..b53561c 100644 --- a/drivers/net/ethernet/intel/ice/ice_adapter.c +++ b/drivers/net/ethernet/intel/ice/ice_adapter.c @@ -13,16 +13,45 @@ static DEFINE_XARRAY(ice_adapters); static DEFINE_MUTEX(ice_adapters_mutex); -static unsigned long ice_adapter_index(u64 dsn) +#define ICE_ADAPTER_FIXED_INDEX BIT_ULL(63) + +#define ICE_ADAPTER_INDEX_E825C \ + (ICE_DEV_ID_E825C_BACKPLANE | ICE_ADAPTER_FIXED_INDEX) + +static u64 ice_adapter_index(struct pci_dev *pdev) { + switch (pdev->device) { + case ICE_DEV_ID_E825C_BACKPLANE: + case ICE_DEV_ID_E825C_QSFP: + case ICE_DEV_ID_E825C_SFP: + case ICE_DEV_ID_E825C_SGMII: + /* E825C devices have multiple NACs which are connected to the + * same clock source, and which must share the same + * ice_adapter structure. We can't use the serial number since + * each NAC has its own NVM generated with its own unique + * Device Serial Number. Instead, rely on the embedded nature + * of the E825C devices, and use a fixed index. This relies on + * the fact that all E825C physical functions in a given + * system are part of the same overall device. + */ + return ICE_ADAPTER_INDEX_E825C; + default: + return pci_get_dsn(pdev) & ~ICE_ADAPTER_FIXED_INDEX; + } +} + +static unsigned long ice_adapter_xa_index(struct pci_dev *pdev) +{ + u64 index = ice_adapter_index(pdev); + #if BITS_PER_LONG == 64 - return dsn; + return index; #else - return (u32)dsn ^ (u32)(dsn >> 32); + return (u32)index ^ (u32)(index >> 32); #endif } -static struct ice_adapter *ice_adapter_new(u64 dsn) +static struct ice_adapter *ice_adapter_new(struct pci_dev *pdev) { struct ice_adapter *adapter; @@ -30,7 +59,7 @@ static struct ice_adapter *ice_adapter_new(u64 dsn) if (!adapter) return NULL; - adapter->device_serial_number = dsn; + adapter->index = ice_adapter_index(pdev); spin_lock_init(&adapter->ptp_gltsyn_time_lock); spin_lock_init(&adapter->txq_ctx_lock); refcount_set(&adapter->refcount, 1); @@ -64,24 +93,23 @@ static void ice_adapter_free(struct ice_adapter *adapter) */ struct ice_adapter *ice_adapter_get(struct pci_dev *pdev) { - u64 dsn = pci_get_dsn(pdev); struct ice_adapter *adapter; unsigned long index; int err; - index = ice_adapter_index(dsn); + index = ice_adapter_xa_index(pdev); scoped_guard(mutex, &ice_adapters_mutex) { err = xa_insert(&ice_adapters, index, NULL, GFP_KERNEL); if (err == -EBUSY) { adapter = xa_load(&ice_adapters, index); refcount_inc(&adapter->refcount); - WARN_ON_ONCE(adapter->device_serial_number != dsn); + WARN_ON_ONCE(adapter->index != ice_adapter_index(pdev)); return adapter; } if (err) return ERR_PTR(err); - adapter = ice_adapter_new(dsn); + adapter = ice_adapter_new(pdev); if (!adapter) return ERR_PTR(-ENOMEM); xa_store(&ice_adapters, index, adapter, GFP_KERNEL); @@ -100,11 +128,10 @@ struct ice_adapter *ice_adapter_get(struct pci_dev *pdev) */ void ice_adapter_put(struct pci_dev *pdev) { - u64 dsn = pci_get_dsn(pdev); struct ice_adapter *adapter; unsigned long index; - index = ice_adapter_index(dsn); + index = ice_adapter_xa_index(pdev); scoped_guard(mutex, &ice_adapters_mutex) { adapter = xa_load(&ice_adapters, index); if (WARN_ON(!adapter)) diff --git a/drivers/net/ethernet/intel/ice/ice_adapter.h b/drivers/net/ethernet/intel/ice/ice_adapter.h index db66d03..e95266c 100644 --- a/drivers/net/ethernet/intel/ice/ice_adapter.h +++ b/drivers/net/ethernet/intel/ice/ice_adapter.h @@ -33,7 +33,7 @@ struct ice_port_list { * @txq_ctx_lock: Spinlock protecting access to the GLCOMM_QTX_CNTX_CTL register * @ctrl_pf: Control PF of the adapter * @ports: Ports list - * @device_serial_number: DSN cached for collision detection on 32bit systems + * @index: 64-bit index cached for collision detection on 32bit systems */ struct ice_adapter { refcount_t refcount; @@ -44,7 +44,7 @@ struct ice_adapter { struct ice_pf *ctrl_pf; struct ice_port_list ports; - u64 device_serial_number; + u64 index; }; struct ice_adapter *ice_adapter_get(struct pci_dev *pdev); diff --git a/drivers/net/ethernet/intel/ice/ice_ddp.c b/drivers/net/ethernet/intel/ice/ice_ddp.c index e2a036c..3b2d9c4 100644 --- a/drivers/net/ethernet/intel/ice/ice_ddp.c +++ b/drivers/net/ethernet/intel/ice/ice_ddp.c @@ -2377,7 +2377,13 @@ ice_get_set_tx_topo(struct ice_hw *hw, u8 *buf, u16 buf_size, * The function will apply the new Tx topology from the package buffer * if available. * - * Return: zero when update was successful, negative values otherwise. + * Return: + * * 0 - Successfully applied topology configuration. + * * -EBUSY - Failed to acquire global configuration lock. + * * -EEXIST - Topology configuration has already been applied. + * * -EIO - Unable to apply topology configuration. + * * -ENODEV - Failed to re-initialize device after applying configuration. + * * Other negative error codes indicate unexpected failures. */ int ice_cfg_tx_topo(struct ice_hw *hw, const void *buf, u32 len) { @@ -2410,7 +2416,7 @@ int ice_cfg_tx_topo(struct ice_hw *hw, const void *buf, u32 len) if (status) { ice_debug(hw, ICE_DBG_INIT, "Get current topology is failed\n"); - return status; + return -EIO; } /* Is default topology already applied ? */ @@ -2497,31 +2503,45 @@ update_topo: ICE_GLOBAL_CFG_LOCK_TIMEOUT); if (status) { ice_debug(hw, ICE_DBG_INIT, "Failed to acquire global lock\n"); - return status; + return -EBUSY; } /* Check if reset was triggered already. */ reg = rd32(hw, GLGEN_RSTAT); if (reg & GLGEN_RSTAT_DEVSTATE_M) { - /* Reset is in progress, re-init the HW again */ ice_debug(hw, ICE_DBG_INIT, "Reset is in progress. Layer topology might be applied already\n"); ice_check_reset(hw); - return 0; + /* Reset is in progress, re-init the HW again */ + goto reinit_hw; } /* Set new topology */ status = ice_get_set_tx_topo(hw, new_topo, size, NULL, NULL, true); if (status) { - ice_debug(hw, ICE_DBG_INIT, "Failed setting Tx topology\n"); - return status; + ice_debug(hw, ICE_DBG_INIT, "Failed to set Tx topology, status %pe\n", + ERR_PTR(status)); + /* only report -EIO here as the caller checks the error value + * and reports an informational error message informing that + * the driver failed to program Tx topology. + */ + status = -EIO; } - /* New topology is updated, delay 1 second before issuing the CORER */ + /* Even if Tx topology config failed, we need to CORE reset here to + * clear the global configuration lock. Delay 1 second to allow + * hardware to settle then issue a CORER + */ msleep(1000); ice_reset(hw, ICE_RESET_CORER); - /* CORER will clear the global lock, so no explicit call - * required for release. - */ + ice_check_reset(hw); + +reinit_hw: + /* Since we triggered a CORER, re-initialize hardware */ + ice_deinit_hw(hw); + if (ice_init_hw(hw)) { + ice_debug(hw, ICE_DBG_INIT, "Failed to re-init hardware after setting Tx topology\n"); + return -ENODEV; + } - return 0; + return status; } diff --git a/drivers/net/ethernet/intel/ice/ice_idc.c b/drivers/net/ethernet/intel/ice/ice_idc.c index 6ab53e4..420d45c 100644 --- a/drivers/net/ethernet/intel/ice/ice_idc.c +++ b/drivers/net/ethernet/intel/ice/ice_idc.c @@ -336,6 +336,7 @@ int ice_plug_aux_dev(struct ice_pf *pf) mutex_lock(&pf->adev_mutex); cdev->adev = adev; mutex_unlock(&pf->adev_mutex); + set_bit(ICE_FLAG_AUX_DEV_CREATED, pf->flags); return 0; } @@ -347,15 +348,16 @@ void ice_unplug_aux_dev(struct ice_pf *pf) { struct auxiliary_device *adev; + if (!test_and_clear_bit(ICE_FLAG_AUX_DEV_CREATED, pf->flags)) + return; + mutex_lock(&pf->adev_mutex); adev = pf->cdev_info->adev; pf->cdev_info->adev = NULL; mutex_unlock(&pf->adev_mutex); - if (adev) { - auxiliary_device_delete(adev); - auxiliary_device_uninit(adev); - } + auxiliary_device_delete(adev); + auxiliary_device_uninit(adev); } /** diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 8e0b06c..cae992d 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -4536,17 +4536,23 @@ ice_init_tx_topology(struct ice_hw *hw, const struct firmware *firmware) dev_info(dev, "Tx scheduling layers switching feature disabled\n"); else dev_info(dev, "Tx scheduling layers switching feature enabled\n"); - /* if there was a change in topology ice_cfg_tx_topo triggered - * a CORER and we need to re-init hw + return 0; + } else if (err == -ENODEV) { + /* If we failed to re-initialize the device, we can no longer + * continue loading. */ - ice_deinit_hw(hw); - err = ice_init_hw(hw); - + dev_warn(dev, "Failed to initialize hardware after applying Tx scheduling configuration.\n"); return err; } else if (err == -EIO) { dev_info(dev, "DDP package does not support Tx scheduling layers switching feature - please update to the latest DDP package and try again\n"); + return 0; + } else if (err == -EEXIST) { + return 0; } + /* Do not treat this as a fatal error. */ + dev_info(dev, "Failed to apply Tx scheduling configuration, err %pe\n", + ERR_PTR(err)); return 0; } diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index 29e0088..d287175 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -1352,7 +1352,7 @@ construct_skb: skb = ice_construct_skb(rx_ring, xdp); /* exit if we failed to retrieve a buffer */ if (!skb) { - rx_ring->ring_stats->rx_stats.alloc_page_failed++; + rx_ring->ring_stats->rx_stats.alloc_buf_failed++; xdp_verdict = ICE_XDP_CONSUMED; } ice_put_rx_mbuf(rx_ring, xdp, &xdp_xmit, ntc, xdp_verdict); diff --git a/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c index 555879b..b19b462 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c +++ b/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c @@ -180,6 +180,58 @@ static int idpf_tx_singleq_csum(struct sk_buff *skb, } /** + * idpf_tx_singleq_dma_map_error - handle TX DMA map errors + * @txq: queue to send buffer on + * @skb: send buffer + * @first: original first buffer info buffer for packet + * @idx: starting point on ring to unwind + */ +static void idpf_tx_singleq_dma_map_error(struct idpf_tx_queue *txq, + struct sk_buff *skb, + struct idpf_tx_buf *first, u16 idx) +{ + struct libeth_sq_napi_stats ss = { }; + struct libeth_cq_pp cp = { + .dev = txq->dev, + .ss = &ss, + }; + + u64_stats_update_begin(&txq->stats_sync); + u64_stats_inc(&txq->q_stats.dma_map_errs); + u64_stats_update_end(&txq->stats_sync); + + /* clear dma mappings for failed tx_buf map */ + for (;;) { + struct idpf_tx_buf *tx_buf; + + tx_buf = &txq->tx_buf[idx]; + libeth_tx_complete(tx_buf, &cp); + if (tx_buf == first) + break; + if (idx == 0) + idx = txq->desc_count; + idx--; + } + + if (skb_is_gso(skb)) { + union idpf_tx_flex_desc *tx_desc; + + /* If we failed a DMA mapping for a TSO packet, we will have + * used one additional descriptor for a context + * descriptor. Reset that here. + */ + tx_desc = &txq->flex_tx[idx]; + memset(tx_desc, 0, sizeof(*tx_desc)); + if (idx == 0) + idx = txq->desc_count; + idx--; + } + + /* Update tail in case netdev_xmit_more was previously true */ + idpf_tx_buf_hw_update(txq, idx, false); +} + +/** * idpf_tx_singleq_map - Build the Tx base descriptor * @tx_q: queue to send buffer on * @first: first buffer info buffer to use @@ -219,8 +271,9 @@ static void idpf_tx_singleq_map(struct idpf_tx_queue *tx_q, for (frag = &skb_shinfo(skb)->frags[0];; frag++) { unsigned int max_data = IDPF_TX_MAX_DESC_DATA_ALIGNED; - if (dma_mapping_error(tx_q->dev, dma)) - return idpf_tx_dma_map_error(tx_q, skb, first, i); + if (unlikely(dma_mapping_error(tx_q->dev, dma))) + return idpf_tx_singleq_dma_map_error(tx_q, skb, + first, i); /* record length, and DMA address */ dma_unmap_len_set(tx_buf, len, size); @@ -362,11 +415,11 @@ netdev_tx_t idpf_tx_singleq_frame(struct sk_buff *skb, { struct idpf_tx_offload_params offload = { }; struct idpf_tx_buf *first; + u32 count, buf_count = 1; int csum, tso, needed; - unsigned int count; __be16 protocol; - count = idpf_tx_desc_count_required(tx_q, skb); + count = idpf_tx_res_count_required(tx_q, skb, &buf_count); if (unlikely(!count)) return idpf_tx_drop_skb(tx_q, skb); diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c index 66a1b04..eaad52a 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c +++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c @@ -8,48 +8,13 @@ #include "idpf_ptp.h" #include "idpf_virtchnl.h" -struct idpf_tx_stash { - struct hlist_node hlist; - struct libeth_sqe buf; -}; - -#define idpf_tx_buf_compl_tag(buf) (*(u32 *)&(buf)->priv) +#define idpf_tx_buf_next(buf) (*(u32 *)&(buf)->priv) LIBETH_SQE_CHECK_PRIV(u32); static bool idpf_chk_linearize(struct sk_buff *skb, unsigned int max_bufs, unsigned int count); /** - * idpf_buf_lifo_push - push a buffer pointer onto stack - * @stack: pointer to stack struct - * @buf: pointer to buf to push - * - * Returns 0 on success, negative on failure - **/ -static int idpf_buf_lifo_push(struct idpf_buf_lifo *stack, - struct idpf_tx_stash *buf) -{ - if (unlikely(stack->top == stack->size)) - return -ENOSPC; - - stack->bufs[stack->top++] = buf; - - return 0; -} - -/** - * idpf_buf_lifo_pop - pop a buffer pointer from stack - * @stack: pointer to stack struct - **/ -static struct idpf_tx_stash *idpf_buf_lifo_pop(struct idpf_buf_lifo *stack) -{ - if (unlikely(!stack->top)) - return NULL; - - return stack->bufs[--stack->top]; -} - -/** * idpf_tx_timeout - Respond to a Tx Hang * @netdev: network interface device structure * @txqueue: TX queue @@ -77,52 +42,22 @@ void idpf_tx_timeout(struct net_device *netdev, unsigned int txqueue) static void idpf_tx_buf_rel_all(struct idpf_tx_queue *txq) { struct libeth_sq_napi_stats ss = { }; - struct idpf_buf_lifo *buf_stack; - struct idpf_tx_stash *stash; struct libeth_cq_pp cp = { .dev = txq->dev, .ss = &ss, }; - struct hlist_node *tmp; - u32 i, tag; + u32 i; /* Buffers already cleared, nothing to do */ if (!txq->tx_buf) return; /* Free all the Tx buffer sk_buffs */ - for (i = 0; i < txq->desc_count; i++) + for (i = 0; i < txq->buf_pool_size; i++) libeth_tx_complete(&txq->tx_buf[i], &cp); kfree(txq->tx_buf); txq->tx_buf = NULL; - - if (!idpf_queue_has(FLOW_SCH_EN, txq)) - return; - - buf_stack = &txq->stash->buf_stack; - if (!buf_stack->bufs) - return; - - /* - * If a Tx timeout occurred, there are potentially still bufs in the - * hash table, free them here. - */ - hash_for_each_safe(txq->stash->sched_buf_hash, tag, tmp, stash, - hlist) { - if (!stash) - continue; - - libeth_tx_complete(&stash->buf, &cp); - hash_del(&stash->hlist); - idpf_buf_lifo_push(buf_stack, stash); - } - - for (i = 0; i < buf_stack->size; i++) - kfree(buf_stack->bufs[i]); - - kfree(buf_stack->bufs); - buf_stack->bufs = NULL; } /** @@ -139,6 +74,9 @@ static void idpf_tx_desc_rel(struct idpf_tx_queue *txq) if (!txq->desc_ring) return; + if (txq->refillq) + kfree(txq->refillq->ring); + dmam_free_coherent(txq->dev, txq->size, txq->desc_ring, txq->dma); txq->desc_ring = NULL; txq->next_to_use = 0; @@ -195,41 +133,18 @@ static void idpf_tx_desc_rel_all(struct idpf_vport *vport) */ static int idpf_tx_buf_alloc_all(struct idpf_tx_queue *tx_q) { - struct idpf_buf_lifo *buf_stack; - int buf_size; - int i; - /* Allocate book keeping buffers only. Buffers to be supplied to HW * are allocated by kernel network stack and received as part of skb */ - buf_size = sizeof(struct idpf_tx_buf) * tx_q->desc_count; - tx_q->tx_buf = kzalloc(buf_size, GFP_KERNEL); + if (idpf_queue_has(FLOW_SCH_EN, tx_q)) + tx_q->buf_pool_size = U16_MAX; + else + tx_q->buf_pool_size = tx_q->desc_count; + tx_q->tx_buf = kcalloc(tx_q->buf_pool_size, sizeof(*tx_q->tx_buf), + GFP_KERNEL); if (!tx_q->tx_buf) return -ENOMEM; - if (!idpf_queue_has(FLOW_SCH_EN, tx_q)) - return 0; - - buf_stack = &tx_q->stash->buf_stack; - - /* Initialize tx buf stack for out-of-order completions if - * flow scheduling offload is enabled - */ - buf_stack->bufs = kcalloc(tx_q->desc_count, sizeof(*buf_stack->bufs), - GFP_KERNEL); - if (!buf_stack->bufs) - return -ENOMEM; - - buf_stack->size = tx_q->desc_count; - buf_stack->top = tx_q->desc_count; - - for (i = 0; i < tx_q->desc_count; i++) { - buf_stack->bufs[i] = kzalloc(sizeof(*buf_stack->bufs[i]), - GFP_KERNEL); - if (!buf_stack->bufs[i]) - return -ENOMEM; - } - return 0; } @@ -244,6 +159,7 @@ static int idpf_tx_desc_alloc(const struct idpf_vport *vport, struct idpf_tx_queue *tx_q) { struct device *dev = tx_q->dev; + struct idpf_sw_queue *refillq; int err; err = idpf_tx_buf_alloc_all(tx_q); @@ -267,6 +183,31 @@ static int idpf_tx_desc_alloc(const struct idpf_vport *vport, tx_q->next_to_clean = 0; idpf_queue_set(GEN_CHK, tx_q); + if (!idpf_queue_has(FLOW_SCH_EN, tx_q)) + return 0; + + refillq = tx_q->refillq; + refillq->desc_count = tx_q->buf_pool_size; + refillq->ring = kcalloc(refillq->desc_count, sizeof(u32), + GFP_KERNEL); + if (!refillq->ring) { + err = -ENOMEM; + goto err_alloc; + } + + for (unsigned int i = 0; i < refillq->desc_count; i++) + refillq->ring[i] = + FIELD_PREP(IDPF_RFL_BI_BUFID_M, i) | + FIELD_PREP(IDPF_RFL_BI_GEN_M, + idpf_queue_has(GEN_CHK, refillq)); + + /* Go ahead and flip the GEN bit since this counts as filling + * up the ring, i.e. we already ring wrapped. + */ + idpf_queue_change(GEN_CHK, refillq); + + tx_q->last_re = tx_q->desc_count - IDPF_TX_SPLITQ_RE_MIN_GAP; + return 0; err_alloc: @@ -317,8 +258,6 @@ static int idpf_tx_desc_alloc_all(struct idpf_vport *vport) for (i = 0; i < vport->num_txq_grp; i++) { for (j = 0; j < vport->txq_grps[i].num_txq; j++) { struct idpf_tx_queue *txq = vport->txq_grps[i].txqs[j]; - u8 gen_bits = 0; - u16 bufidx_mask; err = idpf_tx_desc_alloc(vport, txq); if (err) { @@ -327,34 +266,6 @@ static int idpf_tx_desc_alloc_all(struct idpf_vport *vport) i); goto err_out; } - - if (!idpf_is_queue_model_split(vport->txq_model)) - continue; - - txq->compl_tag_cur_gen = 0; - - /* Determine the number of bits in the bufid - * mask and add one to get the start of the - * generation bits - */ - bufidx_mask = txq->desc_count - 1; - while (bufidx_mask >> 1) { - txq->compl_tag_gen_s++; - bufidx_mask = bufidx_mask >> 1; - } - txq->compl_tag_gen_s++; - - gen_bits = IDPF_TX_SPLITQ_COMPL_TAG_WIDTH - - txq->compl_tag_gen_s; - txq->compl_tag_gen_max = GETMAXVAL(gen_bits); - - /* Set bufid mask based on location of first - * gen bit; it cannot simply be the descriptor - * ring size-1 since we can have size values - * where not all of those bits are set. - */ - txq->compl_tag_bufid_m = - GETMAXVAL(txq->compl_tag_gen_s); } if (!idpf_is_queue_model_split(vport->txq_model)) @@ -603,18 +514,18 @@ static int idpf_rx_hdr_buf_alloc_all(struct idpf_buf_queue *bufq) } /** - * idpf_rx_post_buf_refill - Post buffer id to refill queue + * idpf_post_buf_refill - Post buffer id to refill queue * @refillq: refill queue to post to * @buf_id: buffer id to post */ -static void idpf_rx_post_buf_refill(struct idpf_sw_queue *refillq, u16 buf_id) +static void idpf_post_buf_refill(struct idpf_sw_queue *refillq, u16 buf_id) { u32 nta = refillq->next_to_use; /* store the buffer ID and the SW maintained GEN bit to the refillq */ refillq->ring[nta] = - FIELD_PREP(IDPF_RX_BI_BUFID_M, buf_id) | - FIELD_PREP(IDPF_RX_BI_GEN_M, + FIELD_PREP(IDPF_RFL_BI_BUFID_M, buf_id) | + FIELD_PREP(IDPF_RFL_BI_GEN_M, idpf_queue_has(GEN_CHK, refillq)); if (unlikely(++nta == refillq->desc_count)) { @@ -995,6 +906,11 @@ static void idpf_txq_group_rel(struct idpf_vport *vport) struct idpf_txq_group *txq_grp = &vport->txq_grps[i]; for (j = 0; j < txq_grp->num_txq; j++) { + if (flow_sch_en) { + kfree(txq_grp->txqs[j]->refillq); + txq_grp->txqs[j]->refillq = NULL; + } + kfree(txq_grp->txqs[j]); txq_grp->txqs[j] = NULL; } @@ -1004,9 +920,6 @@ static void idpf_txq_group_rel(struct idpf_vport *vport) kfree(txq_grp->complq); txq_grp->complq = NULL; - - if (flow_sch_en) - kfree(txq_grp->stashes); } kfree(vport->txq_grps); vport->txq_grps = NULL; @@ -1367,7 +1280,6 @@ static int idpf_txq_group_alloc(struct idpf_vport *vport, u16 num_txq) for (i = 0; i < vport->num_txq_grp; i++) { struct idpf_txq_group *tx_qgrp = &vport->txq_grps[i]; struct idpf_adapter *adapter = vport->adapter; - struct idpf_txq_stash *stashes; int j; tx_qgrp->vport = vport; @@ -1380,15 +1292,6 @@ static int idpf_txq_group_alloc(struct idpf_vport *vport, u16 num_txq) goto err_alloc; } - if (split && flow_sch_en) { - stashes = kcalloc(num_txq, sizeof(*stashes), - GFP_KERNEL); - if (!stashes) - goto err_alloc; - - tx_qgrp->stashes = stashes; - } - for (j = 0; j < tx_qgrp->num_txq; j++) { struct idpf_tx_queue *q = tx_qgrp->txqs[j]; @@ -1408,12 +1311,14 @@ static int idpf_txq_group_alloc(struct idpf_vport *vport, u16 num_txq) if (!flow_sch_en) continue; - if (split) { - q->stash = &stashes[j]; - hash_init(q->stash->sched_buf_hash); - } - idpf_queue_set(FLOW_SCH_EN, q); + + q->refillq = kzalloc(sizeof(*q->refillq), GFP_KERNEL); + if (!q->refillq) + goto err_alloc; + + idpf_queue_set(GEN_CHK, q->refillq); + idpf_queue_set(RFL_GEN_CHK, q->refillq); } if (!split) @@ -1697,87 +1602,6 @@ static void idpf_tx_read_tstamp(struct idpf_tx_queue *txq, struct sk_buff *skb) spin_unlock_bh(&tx_tstamp_caps->status_lock); } -/** - * idpf_tx_clean_stashed_bufs - clean bufs that were stored for - * out of order completions - * @txq: queue to clean - * @compl_tag: completion tag of packet to clean (from completion descriptor) - * @cleaned: pointer to stats struct to track cleaned packets/bytes - * @budget: Used to determine if we are in netpoll - */ -static void idpf_tx_clean_stashed_bufs(struct idpf_tx_queue *txq, - u16 compl_tag, - struct libeth_sq_napi_stats *cleaned, - int budget) -{ - struct idpf_tx_stash *stash; - struct hlist_node *tmp_buf; - struct libeth_cq_pp cp = { - .dev = txq->dev, - .ss = cleaned, - .napi = budget, - }; - - /* Buffer completion */ - hash_for_each_possible_safe(txq->stash->sched_buf_hash, stash, tmp_buf, - hlist, compl_tag) { - if (unlikely(idpf_tx_buf_compl_tag(&stash->buf) != compl_tag)) - continue; - - hash_del(&stash->hlist); - - if (stash->buf.type == LIBETH_SQE_SKB && - (skb_shinfo(stash->buf.skb)->tx_flags & SKBTX_IN_PROGRESS)) - idpf_tx_read_tstamp(txq, stash->buf.skb); - - libeth_tx_complete(&stash->buf, &cp); - - /* Push shadow buf back onto stack */ - idpf_buf_lifo_push(&txq->stash->buf_stack, stash); - } -} - -/** - * idpf_stash_flow_sch_buffers - store buffer parameters info to be freed at a - * later time (only relevant for flow scheduling mode) - * @txq: Tx queue to clean - * @tx_buf: buffer to store - */ -static int idpf_stash_flow_sch_buffers(struct idpf_tx_queue *txq, - struct idpf_tx_buf *tx_buf) -{ - struct idpf_tx_stash *stash; - - if (unlikely(tx_buf->type <= LIBETH_SQE_CTX)) - return 0; - - stash = idpf_buf_lifo_pop(&txq->stash->buf_stack); - if (unlikely(!stash)) { - net_err_ratelimited("%s: No out-of-order TX buffers left!\n", - netdev_name(txq->netdev)); - - return -ENOMEM; - } - - /* Store buffer params in shadow buffer */ - stash->buf.skb = tx_buf->skb; - stash->buf.bytes = tx_buf->bytes; - stash->buf.packets = tx_buf->packets; - stash->buf.type = tx_buf->type; - stash->buf.nr_frags = tx_buf->nr_frags; - dma_unmap_addr_set(&stash->buf, dma, dma_unmap_addr(tx_buf, dma)); - dma_unmap_len_set(&stash->buf, len, dma_unmap_len(tx_buf, len)); - idpf_tx_buf_compl_tag(&stash->buf) = idpf_tx_buf_compl_tag(tx_buf); - - /* Add buffer to buf_hash table to be freed later */ - hash_add(txq->stash->sched_buf_hash, &stash->hlist, - idpf_tx_buf_compl_tag(&stash->buf)); - - tx_buf->type = LIBETH_SQE_EMPTY; - - return 0; -} - #define idpf_tx_splitq_clean_bump_ntc(txq, ntc, desc, buf) \ do { \ if (unlikely(++(ntc) == (txq)->desc_count)) { \ @@ -1805,14 +1629,8 @@ do { \ * Separate packet completion events will be reported on the completion queue, * and the buffers will be cleaned separately. The stats are not updated from * this function when using flow-based scheduling. - * - * Furthermore, in flow scheduling mode, check to make sure there are enough - * reserve buffers to stash the packet. If there are not, return early, which - * will leave next_to_clean pointing to the packet that failed to be stashed. - * - * Return: false in the scenario above, true otherwise. */ -static bool idpf_tx_splitq_clean(struct idpf_tx_queue *tx_q, u16 end, +static void idpf_tx_splitq_clean(struct idpf_tx_queue *tx_q, u16 end, int napi_budget, struct libeth_sq_napi_stats *cleaned, bool descs_only) @@ -1826,7 +1644,12 @@ static bool idpf_tx_splitq_clean(struct idpf_tx_queue *tx_q, u16 end, .napi = napi_budget, }; struct idpf_tx_buf *tx_buf; - bool clean_complete = true; + + if (descs_only) { + /* Bump ring index to mark as cleaned. */ + tx_q->next_to_clean = end; + return; + } tx_desc = &tx_q->flex_tx[ntc]; next_pending_desc = &tx_q->flex_tx[end]; @@ -1846,136 +1669,61 @@ static bool idpf_tx_splitq_clean(struct idpf_tx_queue *tx_q, u16 end, break; eop_idx = tx_buf->rs_idx; + libeth_tx_complete(tx_buf, &cp); - if (descs_only) { - if (IDPF_TX_BUF_RSV_UNUSED(tx_q) < tx_buf->nr_frags) { - clean_complete = false; - goto tx_splitq_clean_out; - } - - idpf_stash_flow_sch_buffers(tx_q, tx_buf); + /* unmap remaining buffers */ + while (ntc != eop_idx) { + idpf_tx_splitq_clean_bump_ntc(tx_q, ntc, + tx_desc, tx_buf); - while (ntc != eop_idx) { - idpf_tx_splitq_clean_bump_ntc(tx_q, ntc, - tx_desc, tx_buf); - idpf_stash_flow_sch_buffers(tx_q, tx_buf); - } - } else { + /* unmap any remaining paged data */ libeth_tx_complete(tx_buf, &cp); - - /* unmap remaining buffers */ - while (ntc != eop_idx) { - idpf_tx_splitq_clean_bump_ntc(tx_q, ntc, - tx_desc, tx_buf); - - /* unmap any remaining paged data */ - libeth_tx_complete(tx_buf, &cp); - } } fetch_next_txq_desc: idpf_tx_splitq_clean_bump_ntc(tx_q, ntc, tx_desc, tx_buf); } -tx_splitq_clean_out: tx_q->next_to_clean = ntc; - - return clean_complete; } -#define idpf_tx_clean_buf_ring_bump_ntc(txq, ntc, buf) \ -do { \ - (buf)++; \ - (ntc)++; \ - if (unlikely((ntc) == (txq)->desc_count)) { \ - buf = (txq)->tx_buf; \ - ntc = 0; \ - } \ -} while (0) - /** - * idpf_tx_clean_buf_ring - clean flow scheduling TX queue buffers + * idpf_tx_clean_bufs - clean flow scheduling TX queue buffers * @txq: queue to clean - * @compl_tag: completion tag of packet to clean (from completion descriptor) + * @buf_id: packet's starting buffer ID, from completion descriptor * @cleaned: pointer to stats struct to track cleaned packets/bytes * @budget: Used to determine if we are in netpoll * - * Cleans all buffers associated with the input completion tag either from the - * TX buffer ring or from the hash table if the buffers were previously - * stashed. Returns the byte/segment count for the cleaned packet associated - * this completion tag. + * Clean all buffers associated with the packet starting at buf_id. Returns the + * byte/segment count for the cleaned packet. */ -static bool idpf_tx_clean_buf_ring(struct idpf_tx_queue *txq, u16 compl_tag, - struct libeth_sq_napi_stats *cleaned, - int budget) +static void idpf_tx_clean_bufs(struct idpf_tx_queue *txq, u32 buf_id, + struct libeth_sq_napi_stats *cleaned, + int budget) { - u16 idx = compl_tag & txq->compl_tag_bufid_m; struct idpf_tx_buf *tx_buf = NULL; struct libeth_cq_pp cp = { .dev = txq->dev, .ss = cleaned, .napi = budget, }; - u16 ntc, orig_idx = idx; - - tx_buf = &txq->tx_buf[idx]; - - if (unlikely(tx_buf->type <= LIBETH_SQE_CTX || - idpf_tx_buf_compl_tag(tx_buf) != compl_tag)) - return false; + tx_buf = &txq->tx_buf[buf_id]; if (tx_buf->type == LIBETH_SQE_SKB) { if (skb_shinfo(tx_buf->skb)->tx_flags & SKBTX_IN_PROGRESS) idpf_tx_read_tstamp(txq, tx_buf->skb); libeth_tx_complete(tx_buf, &cp); + idpf_post_buf_refill(txq->refillq, buf_id); } - idpf_tx_clean_buf_ring_bump_ntc(txq, idx, tx_buf); + while (idpf_tx_buf_next(tx_buf) != IDPF_TXBUF_NULL) { + buf_id = idpf_tx_buf_next(tx_buf); - while (idpf_tx_buf_compl_tag(tx_buf) == compl_tag) { + tx_buf = &txq->tx_buf[buf_id]; libeth_tx_complete(tx_buf, &cp); - idpf_tx_clean_buf_ring_bump_ntc(txq, idx, tx_buf); + idpf_post_buf_refill(txq->refillq, buf_id); } - - /* - * It's possible the packet we just cleaned was an out of order - * completion, which means we can stash the buffers starting from - * the original next_to_clean and reuse the descriptors. We need - * to compare the descriptor ring next_to_clean packet's "first" buffer - * to the "first" buffer of the packet we just cleaned to determine if - * this is the case. Howevever, next_to_clean can point to either a - * reserved buffer that corresponds to a context descriptor used for the - * next_to_clean packet (TSO packet) or the "first" buffer (single - * packet). The orig_idx from the packet we just cleaned will always - * point to the "first" buffer. If next_to_clean points to a reserved - * buffer, let's bump ntc once and start the comparison from there. - */ - ntc = txq->next_to_clean; - tx_buf = &txq->tx_buf[ntc]; - - if (tx_buf->type == LIBETH_SQE_CTX) - idpf_tx_clean_buf_ring_bump_ntc(txq, ntc, tx_buf); - - /* - * If ntc still points to a different "first" buffer, clean the - * descriptor ring and stash all of the buffers for later cleaning. If - * we cannot stash all of the buffers, next_to_clean will point to the - * "first" buffer of the packet that could not be stashed and cleaning - * will start there next time. - */ - if (unlikely(tx_buf != &txq->tx_buf[orig_idx] && - !idpf_tx_splitq_clean(txq, orig_idx, budget, cleaned, - true))) - return true; - - /* - * Otherwise, update next_to_clean to reflect the cleaning that was - * done above. - */ - txq->next_to_clean = idx; - - return true; } /** @@ -1994,22 +1742,17 @@ static void idpf_tx_handle_rs_completion(struct idpf_tx_queue *txq, struct libeth_sq_napi_stats *cleaned, int budget) { - u16 compl_tag; + /* RS completion contains queue head for queue based scheduling or + * completion tag for flow based scheduling. + */ + u16 rs_compl_val = le16_to_cpu(desc->q_head_compl_tag.q_head); if (!idpf_queue_has(FLOW_SCH_EN, txq)) { - u16 head = le16_to_cpu(desc->q_head_compl_tag.q_head); - - idpf_tx_splitq_clean(txq, head, budget, cleaned, false); + idpf_tx_splitq_clean(txq, rs_compl_val, budget, cleaned, false); return; } - compl_tag = le16_to_cpu(desc->q_head_compl_tag.compl_tag); - - /* If we didn't clean anything on the ring, this packet must be - * in the hash table. Go clean it there. - */ - if (!idpf_tx_clean_buf_ring(txq, compl_tag, cleaned, budget)) - idpf_tx_clean_stashed_bufs(txq, compl_tag, cleaned, budget); + idpf_tx_clean_bufs(txq, rs_compl_val, cleaned, budget); } /** @@ -2126,8 +1869,7 @@ fetch_next_desc: /* Update BQL */ nq = netdev_get_tx_queue(tx_q->netdev, tx_q->idx); - dont_wake = !complq_ok || IDPF_TX_BUF_RSV_LOW(tx_q) || - np->state != __IDPF_VPORT_UP || + dont_wake = !complq_ok || np->state != __IDPF_VPORT_UP || !netif_carrier_ok(tx_q->netdev); /* Check if the TXQ needs to and can be restarted */ __netif_txq_completed_wake(nq, tx_q->cleaned_pkts, tx_q->cleaned_bytes, @@ -2184,15 +1926,21 @@ void idpf_tx_splitq_build_flow_desc(union idpf_tx_flex_desc *desc, desc->flow.qw1.compl_tag = cpu_to_le16(params->compl_tag); } -/* Global conditions to tell whether the txq (and related resources) - * has room to allow the use of "size" descriptors. +/** + * idpf_tx_splitq_has_room - check if enough Tx splitq resources are available + * @tx_q: the queue to be checked + * @descs_needed: number of descriptors required for this packet + * @bufs_needed: number of Tx buffers required for this packet + * + * Return: 0 if no room available, 1 otherwise */ -static int idpf_txq_has_room(struct idpf_tx_queue *tx_q, u32 size) +static int idpf_txq_has_room(struct idpf_tx_queue *tx_q, u32 descs_needed, + u32 bufs_needed) { - if (IDPF_DESC_UNUSED(tx_q) < size || + if (IDPF_DESC_UNUSED(tx_q) < descs_needed || IDPF_TX_COMPLQ_PENDING(tx_q->txq_grp) > IDPF_TX_COMPLQ_OVERFLOW_THRESH(tx_q->txq_grp->complq) || - IDPF_TX_BUF_RSV_LOW(tx_q)) + idpf_tx_splitq_get_free_bufs(tx_q->refillq) < bufs_needed) return 0; return 1; } @@ -2201,14 +1949,21 @@ static int idpf_txq_has_room(struct idpf_tx_queue *tx_q, u32 size) * idpf_tx_maybe_stop_splitq - 1st level check for Tx splitq stop conditions * @tx_q: the queue to be checked * @descs_needed: number of descriptors required for this packet + * @bufs_needed: number of buffers needed for this packet * - * Returns 0 if stop is not needed + * Return: 0 if stop is not needed */ static int idpf_tx_maybe_stop_splitq(struct idpf_tx_queue *tx_q, - unsigned int descs_needed) + u32 descs_needed, + u32 bufs_needed) { + /* Since we have multiple resources to check for splitq, our + * start,stop_thrs becomes a boolean check instead of a count + * threshold. + */ if (netif_subqueue_maybe_stop(tx_q->netdev, tx_q->idx, - idpf_txq_has_room(tx_q, descs_needed), + idpf_txq_has_room(tx_q, descs_needed, + bufs_needed), 1, 1)) return 0; @@ -2250,14 +2005,16 @@ void idpf_tx_buf_hw_update(struct idpf_tx_queue *tx_q, u32 val, } /** - * idpf_tx_desc_count_required - calculate number of Tx descriptors needed + * idpf_tx_res_count_required - get number of Tx resources needed for this pkt * @txq: queue to send buffer on * @skb: send buffer + * @bufs_needed: (output) number of buffers needed for this skb. * - * Returns number of data descriptors needed for this skb. + * Return: number of data descriptors and buffers needed for this skb. */ -unsigned int idpf_tx_desc_count_required(struct idpf_tx_queue *txq, - struct sk_buff *skb) +unsigned int idpf_tx_res_count_required(struct idpf_tx_queue *txq, + struct sk_buff *skb, + u32 *bufs_needed) { const struct skb_shared_info *shinfo; unsigned int count = 0, i; @@ -2268,6 +2025,7 @@ unsigned int idpf_tx_desc_count_required(struct idpf_tx_queue *txq, return count; shinfo = skb_shinfo(skb); + *bufs_needed += shinfo->nr_frags; for (i = 0; i < shinfo->nr_frags; i++) { unsigned int size; @@ -2297,71 +2055,89 @@ unsigned int idpf_tx_desc_count_required(struct idpf_tx_queue *txq, } /** - * idpf_tx_dma_map_error - handle TX DMA map errors - * @txq: queue to send buffer on - * @skb: send buffer - * @first: original first buffer info buffer for packet - * @idx: starting point on ring to unwind + * idpf_tx_splitq_bump_ntu - adjust NTU and generation + * @txq: the tx ring to wrap + * @ntu: ring index to bump */ -void idpf_tx_dma_map_error(struct idpf_tx_queue *txq, struct sk_buff *skb, - struct idpf_tx_buf *first, u16 idx) +static unsigned int idpf_tx_splitq_bump_ntu(struct idpf_tx_queue *txq, u16 ntu) { - struct libeth_sq_napi_stats ss = { }; - struct libeth_cq_pp cp = { - .dev = txq->dev, - .ss = &ss, - }; + ntu++; - u64_stats_update_begin(&txq->stats_sync); - u64_stats_inc(&txq->q_stats.dma_map_errs); - u64_stats_update_end(&txq->stats_sync); + if (ntu == txq->desc_count) + ntu = 0; - /* clear dma mappings for failed tx_buf map */ - for (;;) { - struct idpf_tx_buf *tx_buf; + return ntu; +} - tx_buf = &txq->tx_buf[idx]; - libeth_tx_complete(tx_buf, &cp); - if (tx_buf == first) - break; - if (idx == 0) - idx = txq->desc_count; - idx--; - } +/** + * idpf_tx_get_free_buf_id - get a free buffer ID from the refill queue + * @refillq: refill queue to get buffer ID from + * @buf_id: return buffer ID + * + * Return: true if a buffer ID was found, false if not + */ +static bool idpf_tx_get_free_buf_id(struct idpf_sw_queue *refillq, + u32 *buf_id) +{ + u32 ntc = refillq->next_to_clean; + u32 refill_desc; - if (skb_is_gso(skb)) { - union idpf_tx_flex_desc *tx_desc; + refill_desc = refillq->ring[ntc]; - /* If we failed a DMA mapping for a TSO packet, we will have - * used one additional descriptor for a context - * descriptor. Reset that here. - */ - tx_desc = &txq->flex_tx[idx]; - memset(tx_desc, 0, sizeof(*tx_desc)); - if (idx == 0) - idx = txq->desc_count; - idx--; + if (unlikely(idpf_queue_has(RFL_GEN_CHK, refillq) != + !!(refill_desc & IDPF_RFL_BI_GEN_M))) + return false; + + *buf_id = FIELD_GET(IDPF_RFL_BI_BUFID_M, refill_desc); + + if (unlikely(++ntc == refillq->desc_count)) { + idpf_queue_change(RFL_GEN_CHK, refillq); + ntc = 0; } - /* Update tail in case netdev_xmit_more was previously true */ - idpf_tx_buf_hw_update(txq, idx, false); + refillq->next_to_clean = ntc; + + return true; } /** - * idpf_tx_splitq_bump_ntu - adjust NTU and generation - * @txq: the tx ring to wrap - * @ntu: ring index to bump + * idpf_tx_splitq_pkt_err_unmap - Unmap buffers and bump tail in case of error + * @txq: Tx queue to unwind + * @params: pointer to splitq params struct + * @first: starting buffer for packet to unmap */ -static unsigned int idpf_tx_splitq_bump_ntu(struct idpf_tx_queue *txq, u16 ntu) +static void idpf_tx_splitq_pkt_err_unmap(struct idpf_tx_queue *txq, + struct idpf_tx_splitq_params *params, + struct idpf_tx_buf *first) { - ntu++; + struct idpf_sw_queue *refillq = txq->refillq; + struct libeth_sq_napi_stats ss = { }; + struct idpf_tx_buf *tx_buf = first; + struct libeth_cq_pp cp = { + .dev = txq->dev, + .ss = &ss, + }; - if (ntu == txq->desc_count) { - ntu = 0; - txq->compl_tag_cur_gen = IDPF_TX_ADJ_COMPL_TAG_GEN(txq); + u64_stats_update_begin(&txq->stats_sync); + u64_stats_inc(&txq->q_stats.dma_map_errs); + u64_stats_update_end(&txq->stats_sync); + + libeth_tx_complete(tx_buf, &cp); + while (idpf_tx_buf_next(tx_buf) != IDPF_TXBUF_NULL) { + tx_buf = &txq->tx_buf[idpf_tx_buf_next(tx_buf)]; + libeth_tx_complete(tx_buf, &cp); } - return ntu; + /* Update tail in case netdev_xmit_more was previously true. */ + idpf_tx_buf_hw_update(txq, params->prev_ntu, false); + + if (!refillq) + return; + + /* Restore refillq state to avoid leaking tags. */ + if (params->prev_refill_gen != idpf_queue_has(RFL_GEN_CHK, refillq)) + idpf_queue_change(RFL_GEN_CHK, refillq); + refillq->next_to_clean = params->prev_refill_ntc; } /** @@ -2385,6 +2161,7 @@ static void idpf_tx_splitq_map(struct idpf_tx_queue *tx_q, struct netdev_queue *nq; struct sk_buff *skb; skb_frag_t *frag; + u32 next_buf_id; u16 td_cmd = 0; dma_addr_t dma; @@ -2402,17 +2179,16 @@ static void idpf_tx_splitq_map(struct idpf_tx_queue *tx_q, tx_buf = first; first->nr_frags = 0; - params->compl_tag = - (tx_q->compl_tag_cur_gen << tx_q->compl_tag_gen_s) | i; - for (frag = &skb_shinfo(skb)->frags[0];; frag++) { unsigned int max_data = IDPF_TX_MAX_DESC_DATA_ALIGNED; - if (dma_mapping_error(tx_q->dev, dma)) - return idpf_tx_dma_map_error(tx_q, skb, first, i); + if (unlikely(dma_mapping_error(tx_q->dev, dma))) { + idpf_tx_buf_next(tx_buf) = IDPF_TXBUF_NULL; + return idpf_tx_splitq_pkt_err_unmap(tx_q, params, + first); + } first->nr_frags++; - idpf_tx_buf_compl_tag(tx_buf) = params->compl_tag; tx_buf->type = LIBETH_SQE_FRAG; /* record length, and DMA address */ @@ -2468,29 +2244,12 @@ static void idpf_tx_splitq_map(struct idpf_tx_queue *tx_q, max_data); if (unlikely(++i == tx_q->desc_count)) { - tx_buf = tx_q->tx_buf; tx_desc = &tx_q->flex_tx[0]; i = 0; - tx_q->compl_tag_cur_gen = - IDPF_TX_ADJ_COMPL_TAG_GEN(tx_q); } else { - tx_buf++; tx_desc++; } - /* Since this packet has a buffer that is going to span - * multiple descriptors, it's going to leave holes in - * to the TX buffer ring. To ensure these holes do not - * cause issues in the cleaning routines, we will clear - * them of any stale data and assign them the same - * completion tag as the current packet. Then when the - * packet is being cleaned, the cleaning routines will - * simply pass over these holes and finish cleaning the - * rest of the packet. - */ - tx_buf->type = LIBETH_SQE_EMPTY; - idpf_tx_buf_compl_tag(tx_buf) = params->compl_tag; - /* Adjust the DMA offset and the remaining size of the * fragment. On the first iteration of this loop, * max_data will be >= 12K and <= 16K-1. On any @@ -2515,15 +2274,25 @@ static void idpf_tx_splitq_map(struct idpf_tx_queue *tx_q, idpf_tx_splitq_build_desc(tx_desc, params, td_cmd, size); if (unlikely(++i == tx_q->desc_count)) { - tx_buf = tx_q->tx_buf; tx_desc = &tx_q->flex_tx[0]; i = 0; - tx_q->compl_tag_cur_gen = IDPF_TX_ADJ_COMPL_TAG_GEN(tx_q); } else { - tx_buf++; tx_desc++; } + if (idpf_queue_has(FLOW_SCH_EN, tx_q)) { + if (unlikely(!idpf_tx_get_free_buf_id(tx_q->refillq, + &next_buf_id))) { + idpf_tx_buf_next(tx_buf) = IDPF_TXBUF_NULL; + return idpf_tx_splitq_pkt_err_unmap(tx_q, params, + first); + } + } else { + next_buf_id = i; + } + idpf_tx_buf_next(tx_buf) = next_buf_id; + tx_buf = &tx_q->tx_buf[next_buf_id]; + size = skb_frag_size(frag); data_len -= size; @@ -2538,6 +2307,7 @@ static void idpf_tx_splitq_map(struct idpf_tx_queue *tx_q, /* write last descriptor with RS and EOP bits */ first->rs_idx = i; + idpf_tx_buf_next(tx_buf) = IDPF_TXBUF_NULL; td_cmd |= params->eop_cmd; idpf_tx_splitq_build_desc(tx_desc, params, td_cmd, size); i = idpf_tx_splitq_bump_ntu(tx_q, i); @@ -2746,8 +2516,6 @@ idpf_tx_splitq_get_ctx_desc(struct idpf_tx_queue *txq) union idpf_flex_tx_ctx_desc *desc; int i = txq->next_to_use; - txq->tx_buf[i].type = LIBETH_SQE_CTX; - /* grab the next descriptor */ desc = &txq->flex_ctx[i]; txq->next_to_use = idpf_tx_splitq_bump_ntu(txq, i); @@ -2841,6 +2609,21 @@ static void idpf_tx_set_tstamp_desc(union idpf_flex_tx_ctx_desc *ctx_desc, #endif /* CONFIG_PTP_1588_CLOCK */ /** + * idpf_tx_splitq_need_re - check whether RE bit needs to be set + * @tx_q: pointer to Tx queue + * + * Return: true if RE bit needs to be set, false otherwise + */ +static bool idpf_tx_splitq_need_re(struct idpf_tx_queue *tx_q) +{ + int gap = tx_q->next_to_use - tx_q->last_re; + + gap += (gap < 0) ? tx_q->desc_count : 0; + + return gap >= IDPF_TX_SPLITQ_RE_MIN_GAP; +} + +/** * idpf_tx_splitq_frame - Sends buffer on Tx ring using flex descriptors * @skb: send buffer * @tx_q: queue to send buffer on @@ -2850,13 +2633,16 @@ static void idpf_tx_set_tstamp_desc(union idpf_flex_tx_ctx_desc *ctx_desc, static netdev_tx_t idpf_tx_splitq_frame(struct sk_buff *skb, struct idpf_tx_queue *tx_q) { - struct idpf_tx_splitq_params tx_params = { }; + struct idpf_tx_splitq_params tx_params = { + .prev_ntu = tx_q->next_to_use, + }; union idpf_flex_tx_ctx_desc *ctx_desc; struct idpf_tx_buf *first; - unsigned int count; + u32 count, buf_count = 1; int tso, idx; + u32 buf_id; - count = idpf_tx_desc_count_required(tx_q, skb); + count = idpf_tx_res_count_required(tx_q, skb, &buf_count); if (unlikely(!count)) return idpf_tx_drop_skb(tx_q, skb); @@ -2866,7 +2652,7 @@ static netdev_tx_t idpf_tx_splitq_frame(struct sk_buff *skb, /* Check for splitq specific TX resources */ count += (IDPF_TX_DESCS_PER_CACHE_LINE + tso); - if (idpf_tx_maybe_stop_splitq(tx_q, count)) { + if (idpf_tx_maybe_stop_splitq(tx_q, count, buf_count)) { idpf_tx_buf_hw_update(tx_q, tx_q->next_to_use, false); return NETDEV_TX_BUSY; @@ -2898,36 +2684,47 @@ static netdev_tx_t idpf_tx_splitq_frame(struct sk_buff *skb, idpf_tx_set_tstamp_desc(ctx_desc, idx); } - /* record the location of the first descriptor for this packet */ - first = &tx_q->tx_buf[tx_q->next_to_use]; - first->skb = skb; + if (idpf_queue_has(FLOW_SCH_EN, tx_q)) { + struct idpf_sw_queue *refillq = tx_q->refillq; - if (tso) { - first->packets = tx_params.offload.tso_segs; - first->bytes = skb->len + - ((first->packets - 1) * tx_params.offload.tso_hdr_len); - } else { - first->packets = 1; - first->bytes = max_t(unsigned int, skb->len, ETH_ZLEN); - } + /* Save refillq state in case of a packet rollback. Otherwise, + * the tags will be leaked since they will be popped from the + * refillq but never reposted during cleaning. + */ + tx_params.prev_refill_gen = + idpf_queue_has(RFL_GEN_CHK, refillq); + tx_params.prev_refill_ntc = refillq->next_to_clean; + + if (unlikely(!idpf_tx_get_free_buf_id(tx_q->refillq, + &buf_id))) { + if (tx_params.prev_refill_gen != + idpf_queue_has(RFL_GEN_CHK, refillq)) + idpf_queue_change(RFL_GEN_CHK, refillq); + refillq->next_to_clean = tx_params.prev_refill_ntc; + + tx_q->next_to_use = tx_params.prev_ntu; + return idpf_tx_drop_skb(tx_q, skb); + } + tx_params.compl_tag = buf_id; - if (idpf_queue_has(FLOW_SCH_EN, tx_q)) { tx_params.dtype = IDPF_TX_DESC_DTYPE_FLEX_FLOW_SCHE; tx_params.eop_cmd = IDPF_TXD_FLEX_FLOW_CMD_EOP; - /* Set the RE bit to catch any packets that may have not been - * stashed during RS completion cleaning. MIN_GAP is set to - * MIN_RING size to ensure it will be set at least once each - * time around the ring. + /* Set the RE bit to periodically "clean" the descriptor ring. + * MIN_GAP is set to MIN_RING size to ensure it will be set at + * least once each time around the ring. */ - if (!(tx_q->next_to_use % IDPF_TX_SPLITQ_RE_MIN_GAP)) { + if (idpf_tx_splitq_need_re(tx_q)) { tx_params.eop_cmd |= IDPF_TXD_FLEX_FLOW_CMD_RE; tx_q->txq_grp->num_completions_pending++; + tx_q->last_re = tx_q->next_to_use; } if (skb->ip_summed == CHECKSUM_PARTIAL) tx_params.offload.td_cmd |= IDPF_TXD_FLEX_FLOW_CMD_CS_EN; } else { + buf_id = tx_q->next_to_use; + tx_params.dtype = IDPF_TX_DESC_DTYPE_FLEX_L2TAG1_L2TAG2; tx_params.eop_cmd = IDPF_TXD_LAST_DESC_CMD; @@ -2935,6 +2732,18 @@ static netdev_tx_t idpf_tx_splitq_frame(struct sk_buff *skb, tx_params.offload.td_cmd |= IDPF_TX_FLEX_DESC_CMD_CS_EN; } + first = &tx_q->tx_buf[buf_id]; + first->skb = skb; + + if (tso) { + first->packets = tx_params.offload.tso_segs; + first->bytes = skb->len + + ((first->packets - 1) * tx_params.offload.tso_hdr_len); + } else { + first->packets = 1; + first->bytes = max_t(unsigned int, skb->len, ETH_ZLEN); + } + idpf_tx_splitq_map(tx_q, &tx_params, first); return NETDEV_TX_OK; @@ -3472,7 +3281,7 @@ payload: skip_data: rx_buf->netmem = 0; - idpf_rx_post_buf_refill(refillq, buf_id); + idpf_post_buf_refill(refillq, buf_id); IDPF_RX_BUMP_NTC(rxq, ntc); /* skip if it is non EOP desc */ @@ -3580,10 +3389,10 @@ static void idpf_rx_clean_refillq(struct idpf_buf_queue *bufq, bool failure; if (idpf_queue_has(RFL_GEN_CHK, refillq) != - !!(refill_desc & IDPF_RX_BI_GEN_M)) + !!(refill_desc & IDPF_RFL_BI_GEN_M)) break; - buf_id = FIELD_GET(IDPF_RX_BI_BUFID_M, refill_desc); + buf_id = FIELD_GET(IDPF_RFL_BI_BUFID_M, refill_desc); failure = idpf_rx_update_bufq_desc(bufq, buf_id, buf_desc); if (failure) break; diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.h b/drivers/net/ethernet/intel/idpf/idpf_txrx.h index 281de65..52753df 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_txrx.h +++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.h @@ -108,8 +108,8 @@ do { \ */ #define IDPF_TX_SPLITQ_RE_MIN_GAP 64 -#define IDPF_RX_BI_GEN_M BIT(16) -#define IDPF_RX_BI_BUFID_M GENMASK(15, 0) +#define IDPF_RFL_BI_GEN_M BIT(16) +#define IDPF_RFL_BI_BUFID_M GENMASK(15, 0) #define IDPF_RXD_EOF_SPLITQ VIRTCHNL2_RX_FLEX_DESC_ADV_STATUS0_EOF_M #define IDPF_RXD_EOF_SINGLEQ VIRTCHNL2_RX_BASE_DESC_STATUS_EOF_M @@ -118,10 +118,6 @@ do { \ ((((txq)->next_to_clean > (txq)->next_to_use) ? 0 : (txq)->desc_count) + \ (txq)->next_to_clean - (txq)->next_to_use - 1) -#define IDPF_TX_BUF_RSV_UNUSED(txq) ((txq)->stash->buf_stack.top) -#define IDPF_TX_BUF_RSV_LOW(txq) (IDPF_TX_BUF_RSV_UNUSED(txq) < \ - (txq)->desc_count >> 2) - #define IDPF_TX_COMPLQ_OVERFLOW_THRESH(txcq) ((txcq)->desc_count >> 1) /* Determine the absolute number of completions pending, i.e. the number of * completions that are expected to arrive on the TX completion queue. @@ -131,11 +127,7 @@ do { \ 0 : U32_MAX) + \ (txq)->num_completions_pending - (txq)->complq->num_completions) -#define IDPF_TX_SPLITQ_COMPL_TAG_WIDTH 16 -/* Adjust the generation for the completion tag and wrap if necessary */ -#define IDPF_TX_ADJ_COMPL_TAG_GEN(txq) \ - ((++(txq)->compl_tag_cur_gen) >= (txq)->compl_tag_gen_max ? \ - 0 : (txq)->compl_tag_cur_gen) +#define IDPF_TXBUF_NULL U32_MAX #define IDPF_TXD_LAST_DESC_CMD (IDPF_TX_DESC_CMD_EOP | IDPF_TX_DESC_CMD_RS) @@ -153,18 +145,6 @@ union idpf_tx_flex_desc { #define idpf_tx_buf libeth_sqe /** - * struct idpf_buf_lifo - LIFO for managing OOO completions - * @top: Used to know how many buffers are left - * @size: Total size of LIFO - * @bufs: Backing array - */ -struct idpf_buf_lifo { - u16 top; - u16 size; - struct idpf_tx_stash **bufs; -}; - -/** * struct idpf_tx_offload_params - Offload parameters for a given packet * @tx_flags: Feature flags enabled for this packet * @hdr_offsets: Offset parameter for single queue model @@ -196,6 +176,9 @@ struct idpf_tx_offload_params { * @compl_tag: Associated tag for completion * @td_tag: Descriptor tunneling tag * @offload: Offload parameters + * @prev_ntu: stored TxQ next_to_use in case of rollback + * @prev_refill_ntc: stored refillq next_to_clean in case of packet rollback + * @prev_refill_gen: stored refillq generation bit in case of packet rollback */ struct idpf_tx_splitq_params { enum idpf_tx_desc_dtype_value dtype; @@ -206,6 +189,10 @@ struct idpf_tx_splitq_params { }; struct idpf_tx_offload_params offload; + + u16 prev_ntu; + u16 prev_refill_ntc; + bool prev_refill_gen; }; enum idpf_tx_ctx_desc_eipt_offload { @@ -468,17 +455,6 @@ struct idpf_tx_queue_stats { #define IDPF_DIM_DEFAULT_PROFILE_IX 1 /** - * struct idpf_txq_stash - Tx buffer stash for Flow-based scheduling mode - * @buf_stack: Stack of empty buffers to store buffer info for out of order - * buffer completions. See struct idpf_buf_lifo - * @sched_buf_hash: Hash table to store buffers - */ -struct idpf_txq_stash { - struct idpf_buf_lifo buf_stack; - DECLARE_HASHTABLE(sched_buf_hash, 12); -} ____cacheline_aligned; - -/** * struct idpf_rx_queue - software structure representing a receive queue * @rx: universal receive descriptor array * @single_buf: buffer descriptor array in singleq @@ -610,6 +586,8 @@ libeth_cacheline_set_assert(struct idpf_rx_queue, 64, * @netdev: &net_device corresponding to this queue * @next_to_use: Next descriptor to use * @next_to_clean: Next descriptor to clean + * @last_re: last descriptor index that RE bit was set + * @tx_max_bufs: Max buffers that can be transmitted with scatter-gather * @cleaned_bytes: Splitq only, TXQ only: When a TX completion is received on * the TX completion queue, it can be for any TXQ associated * with that completion queue. This means we can clean up to @@ -620,11 +598,7 @@ libeth_cacheline_set_assert(struct idpf_rx_queue, 64, * only once at the end of the cleaning routine. * @clean_budget: singleq only, queue cleaning budget * @cleaned_pkts: Number of packets cleaned for the above said case - * @tx_max_bufs: Max buffers that can be transmitted with scatter-gather - * @stash: Tx buffer stash for Flow-based scheduling mode - * @compl_tag_bufid_m: Completion tag buffer id mask - * @compl_tag_cur_gen: Used to keep track of current completion tag generation - * @compl_tag_gen_max: To determine when compl_tag_cur_gen should be reset + * @refillq: Pointer to refill queue * @cached_tstamp_caps: Tx timestamp capabilities negotiated with the CP * @tstamp_task: Work that handles Tx timestamp read * @stats_sync: See struct u64_stats_sync @@ -633,6 +607,7 @@ libeth_cacheline_set_assert(struct idpf_rx_queue, 64, * @size: Length of descriptor ring in bytes * @dma: Physical address of ring * @q_vector: Backreference to associated vector + * @buf_pool_size: Total number of idpf_tx_buf */ struct idpf_tx_queue { __cacheline_group_begin_aligned(read_mostly); @@ -654,7 +629,6 @@ struct idpf_tx_queue { u16 desc_count; u16 tx_min_pkt_len; - u16 compl_tag_gen_s; struct net_device *netdev; __cacheline_group_end_aligned(read_mostly); @@ -662,6 +636,8 @@ struct idpf_tx_queue { __cacheline_group_begin_aligned(read_write); u16 next_to_use; u16 next_to_clean; + u16 last_re; + u16 tx_max_bufs; union { u32 cleaned_bytes; @@ -669,12 +645,7 @@ struct idpf_tx_queue { }; u16 cleaned_pkts; - u16 tx_max_bufs; - struct idpf_txq_stash *stash; - - u16 compl_tag_bufid_m; - u16 compl_tag_cur_gen; - u16 compl_tag_gen_max; + struct idpf_sw_queue *refillq; struct idpf_ptp_vport_tx_tstamp_caps *cached_tstamp_caps; struct work_struct *tstamp_task; @@ -689,11 +660,12 @@ struct idpf_tx_queue { dma_addr_t dma; struct idpf_q_vector *q_vector; + u32 buf_pool_size; __cacheline_group_end_aligned(cold); }; libeth_cacheline_set_assert(struct idpf_tx_queue, 64, - 112 + sizeof(struct u64_stats_sync), - 24); + 104 + sizeof(struct u64_stats_sync), + 32); /** * struct idpf_buf_queue - software structure representing a buffer queue @@ -903,7 +875,6 @@ struct idpf_rxq_group { * @vport: Vport back pointer * @num_txq: Number of TX queues associated * @txqs: Array of TX queue pointers - * @stashes: array of OOO stashes for the queues * @complq: Associated completion queue pointer, split queue only * @num_completions_pending: Total number of completions pending for the * completion queue, acculumated for all TX queues @@ -918,7 +889,6 @@ struct idpf_txq_group { u16 num_txq; struct idpf_tx_queue *txqs[IDPF_LARGE_MAX_Q]; - struct idpf_txq_stash *stashes; struct idpf_compl_queue *complq; @@ -1011,6 +981,17 @@ static inline void idpf_vport_intr_set_wb_on_itr(struct idpf_q_vector *q_vector) reg->dyn_ctl); } +/** + * idpf_tx_splitq_get_free_bufs - get number of free buf_ids in refillq + * @refillq: pointer to refillq containing buf_ids + */ +static inline u32 idpf_tx_splitq_get_free_bufs(struct idpf_sw_queue *refillq) +{ + return (refillq->next_to_use > refillq->next_to_clean ? + 0 : refillq->desc_count) + + refillq->next_to_use - refillq->next_to_clean - 1; +} + int idpf_vport_singleq_napi_poll(struct napi_struct *napi, int budget); void idpf_vport_init_num_qs(struct idpf_vport *vport, struct virtchnl2_create_vport *vport_msg); @@ -1038,10 +1019,8 @@ void idpf_tx_buf_hw_update(struct idpf_tx_queue *tx_q, u32 val, bool xmit_more); unsigned int idpf_size_to_txd_count(unsigned int size); netdev_tx_t idpf_tx_drop_skb(struct idpf_tx_queue *tx_q, struct sk_buff *skb); -void idpf_tx_dma_map_error(struct idpf_tx_queue *txq, struct sk_buff *skb, - struct idpf_tx_buf *first, u16 ring_idx); -unsigned int idpf_tx_desc_count_required(struct idpf_tx_queue *txq, - struct sk_buff *skb); +unsigned int idpf_tx_res_count_required(struct idpf_tx_queue *txq, + struct sk_buff *skb, u32 *buf_count); void idpf_tx_timeout(struct net_device *netdev, unsigned int txqueue); netdev_tx_t idpf_tx_singleq_frame(struct sk_buff *skb, struct idpf_tx_queue *tx_q); diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_e610.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_e610.c index d741164..bfeef5b 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_e610.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_e610.c @@ -3125,7 +3125,7 @@ static int ixgbe_get_orom_ver_info(struct ixgbe_hw *hw, if (err) return err; - combo_ver = le32_to_cpu(civd.combo_ver); + combo_ver = get_unaligned_le32(&civd.combo_ver); orom->major = (u8)FIELD_GET(IXGBE_OROM_VER_MASK, combo_ver); orom->patch = (u8)FIELD_GET(IXGBE_OROM_VER_PATCH_MASK, combo_ver); diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_type_e610.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_type_e610.h index d2f22d8..ff8d640 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_type_e610.h +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_type_e610.h @@ -932,7 +932,7 @@ struct ixgbe_orom_civd_info { __le32 combo_ver; /* Combo Image Version number */ u8 combo_name_len; /* Length of the unicode combo image version string, max of 32 */ __le16 combo_name[32]; /* Unicode string representing the Combo Image version */ -}; +} __packed; /* Function specific capabilities */ struct ixgbe_hw_func_caps { diff --git a/drivers/net/ethernet/marvell/octeontx2/af/cgx.c b/drivers/net/ethernet/marvell/octeontx2/af/cgx.c index 4ff19a0..0c46ba8 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/cgx.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/cgx.c @@ -1978,6 +1978,13 @@ static int cgx_probe(struct pci_dev *pdev, const struct pci_device_id *id) goto err_release_regions; } + if (!is_cn20k(pdev) && + !is_cgx_mapped_to_nix(pdev->subsystem_device, cgx->cgx_id)) { + dev_notice(dev, "CGX %d not mapped to NIX, skipping probe\n", + cgx->cgx_id); + goto err_release_regions; + } + cgx->lmac_count = cgx->mac_ops->get_nr_lmacs(cgx); if (!cgx->lmac_count) { dev_notice(dev, "CGX %d LMAC count is zero, skipping probe\n", cgx->cgx_id); diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h index 7ee1fde..18c7bb3 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h @@ -783,6 +783,20 @@ static inline bool is_cn10kb(struct rvu *rvu) return false; } +static inline bool is_cgx_mapped_to_nix(unsigned short id, u8 cgx_id) +{ + /* On CNF10KA and CNF10KB silicons only two CGX blocks are connected + * to NIX. + */ + if (id == PCI_SUBSYS_DEVID_CNF10K_A || id == PCI_SUBSYS_DEVID_CNF10K_B) + return cgx_id <= 1; + + return !(cgx_id && !(id == PCI_SUBSYS_DEVID_96XX || + id == PCI_SUBSYS_DEVID_98XX || + id == PCI_SUBSYS_DEVID_CN10K_A || + id == PCI_SUBSYS_DEVID_CN10K_B)); +} + static inline bool is_rvu_npc_hash_extract_en(struct rvu *rvu) { u64 npc_const3; diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c index f674729..aff17c3 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c @@ -124,7 +124,9 @@ void otx2_get_dev_stats(struct otx2_nic *pfvf) dev_stats->rx_ucast_frames; dev_stats->tx_bytes = OTX2_GET_TX_STATS(TX_OCTS); - dev_stats->tx_drops = OTX2_GET_TX_STATS(TX_DROP); + dev_stats->tx_drops = OTX2_GET_TX_STATS(TX_DROP) + + (unsigned long)atomic_long_read(&dev_stats->tx_discards); + dev_stats->tx_bcast_frames = OTX2_GET_TX_STATS(TX_BCAST); dev_stats->tx_mcast_frames = OTX2_GET_TX_STATS(TX_MCAST); dev_stats->tx_ucast_frames = OTX2_GET_TX_STATS(TX_UCAST); diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h index e3765b7..1c8a3c0 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h @@ -153,6 +153,7 @@ struct otx2_dev_stats { u64 tx_bcast_frames; u64 tx_mcast_frames; u64 tx_drops; + atomic_long_t tx_discards; }; /* Driver counted stats */ diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c index b23585c..5027fae 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c @@ -2220,6 +2220,7 @@ static netdev_tx_t otx2_xmit(struct sk_buff *skb, struct net_device *netdev) { struct otx2_nic *pf = netdev_priv(netdev); int qidx = skb_get_queue_mapping(skb); + struct otx2_dev_stats *dev_stats; struct otx2_snd_queue *sq; struct netdev_queue *txq; int sq_idx; @@ -2232,6 +2233,8 @@ static netdev_tx_t otx2_xmit(struct sk_buff *skb, struct net_device *netdev) /* Check for minimum and maximum packet length */ if (skb->len <= ETH_HLEN || (!skb_shinfo(skb)->gso_size && skb->len > pf->tx_max_pktlen)) { + dev_stats = &pf->hw.dev_stats; + atomic_long_inc(&dev_stats->tx_discards); dev_kfree_skb(skb); return NETDEV_TX_OK; } diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c index 5589fcc..7ebb6e6 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c @@ -417,9 +417,19 @@ static netdev_tx_t otx2vf_xmit(struct sk_buff *skb, struct net_device *netdev) { struct otx2_nic *vf = netdev_priv(netdev); int qidx = skb_get_queue_mapping(skb); + struct otx2_dev_stats *dev_stats; struct otx2_snd_queue *sq; struct netdev_queue *txq; + /* Check for minimum and maximum packet length */ + if (skb->len <= ETH_HLEN || + (!skb_shinfo(skb)->gso_size && skb->len > vf->tx_max_pktlen)) { + dev_stats = &vf->hw.dev_stats; + atomic_long_inc(&dev_stats->tx_discards); + dev_kfree_skb(skb); + return NETDEV_TX_OK; + } + sq = &vf->qset.sq[qidx]; txq = netdev_get_tx_queue(netdev, qidx); diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/rep.c b/drivers/net/ethernet/marvell/octeontx2/nic/rep.c index 25af980..b476733 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/rep.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/rep.c @@ -371,7 +371,8 @@ static void rvu_rep_get_stats(struct work_struct *work) stats->rx_mcast_frames = rsp->rx.mcast; stats->tx_bytes = rsp->tx.octs; stats->tx_frames = rsp->tx.ucast + rsp->tx.bcast + rsp->tx.mcast; - stats->tx_drops = rsp->tx.drop; + stats->tx_drops = rsp->tx.drop + + (unsigned long)atomic_long_read(&stats->tx_discards); exit: mutex_unlock(&priv->mbox.lock); } @@ -418,6 +419,16 @@ static netdev_tx_t rvu_rep_xmit(struct sk_buff *skb, struct net_device *dev) struct otx2_nic *pf = rep->mdev; struct otx2_snd_queue *sq; struct netdev_queue *txq; + struct rep_stats *stats; + + /* Check for minimum and maximum packet length */ + if (skb->len <= ETH_HLEN || + (!skb_shinfo(skb)->gso_size && skb->len > pf->tx_max_pktlen)) { + stats = &rep->stats; + atomic_long_inc(&stats->tx_discards); + dev_kfree_skb(skb); + return NETDEV_TX_OK; + } sq = &pf->qset.sq[rep->rep_id]; txq = netdev_get_tx_queue(dev, 0); diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/rep.h b/drivers/net/ethernet/marvell/octeontx2/nic/rep.h index 38446b3..5bc9e2c 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/rep.h +++ b/drivers/net/ethernet/marvell/octeontx2/nic/rep.h @@ -27,6 +27,7 @@ struct rep_stats { u64 tx_bytes; u64 tx_frames; u64 tx_drops; + atomic_long_t tx_discards; }; struct rep_dev { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c index 3ffa3fb..2c0e0c1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c @@ -160,7 +160,7 @@ static int mlx5_devlink_reload_fw_activate(struct devlink *devlink, struct netli if (err) return err; - mlx5_unload_one_devl_locked(dev, true); + mlx5_sync_reset_unload_flow(dev, true); err = mlx5_health_wait_pci_up(dev); if (err) NL_SET_ERR_MSG_MOD(extack, "FW activate aborted, PCI reads fail after reset"); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c b/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c index 3efa8bf..4720523 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c @@ -575,7 +575,6 @@ int mlx5e_port_manual_buffer_config(struct mlx5e_priv *priv, if (err) return err; } - priv->dcbx.xoff = xoff; /* Apply the settings */ if (update_buffer) { @@ -584,6 +583,8 @@ int mlx5e_port_manual_buffer_config(struct mlx5e_priv *priv, return err; } + priv->dcbx.xoff = xoff; + if (update_prio2buffer) err = mlx5e_port_set_priority2buffer(priv->mdev, prio2buffer); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.h b/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.h index f4a19ff..66d276a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.h @@ -66,11 +66,23 @@ struct mlx5e_port_buffer { struct mlx5e_bufferx_reg buffer[MLX5E_MAX_NETWORK_BUFFER]; }; +#ifdef CONFIG_MLX5_CORE_EN_DCB int mlx5e_port_manual_buffer_config(struct mlx5e_priv *priv, u32 change, unsigned int mtu, struct ieee_pfc *pfc, u32 *buffer_size, u8 *prio2buffer); +#else +static inline int +mlx5e_port_manual_buffer_config(struct mlx5e_priv *priv, + u32 change, unsigned int mtu, + void *pfc, + u32 *buffer_size, + u8 *prio2buffer) +{ + return 0; +} +#endif int mlx5e_port_query_buffer(struct mlx5e_priv *priv, struct mlx5e_port_buffer *port_buffer); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 21bb88c..e680673 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -49,6 +49,7 @@ #include "en.h" #include "en/dim.h" #include "en/txrx.h" +#include "en/port_buffer.h" #include "en_tc.h" #include "en_rep.h" #include "en_accel/ipsec.h" @@ -138,6 +139,8 @@ void mlx5e_update_carrier(struct mlx5e_priv *priv) if (up) { netdev_info(priv->netdev, "Link up\n"); netif_carrier_on(priv->netdev); + mlx5e_port_manual_buffer_config(priv, 0, priv->netdev->mtu, + NULL, NULL, NULL); } else { netdev_info(priv->netdev, "Link down\n"); netif_carrier_off(priv->netdev); @@ -3040,9 +3043,11 @@ int mlx5e_set_dev_port_mtu(struct mlx5e_priv *priv) struct mlx5e_params *params = &priv->channels.params; struct net_device *netdev = priv->netdev; struct mlx5_core_dev *mdev = priv->mdev; - u16 mtu; + u16 mtu, prev_mtu; int err; + mlx5e_query_mtu(mdev, params, &prev_mtu); + err = mlx5e_set_mtu(mdev, params, params->sw_mtu); if (err) return err; @@ -3052,6 +3057,18 @@ int mlx5e_set_dev_port_mtu(struct mlx5e_priv *priv) netdev_warn(netdev, "%s: VPort MTU %d is different than netdev mtu %d\n", __func__, mtu, params->sw_mtu); + if (mtu != prev_mtu && MLX5_BUFFER_SUPPORTED(mdev)) { + err = mlx5e_port_manual_buffer_config(priv, 0, mtu, + NULL, NULL, NULL); + if (err) { + netdev_warn(netdev, "%s: Failed to set Xon/Xoff values with MTU %d (err %d), setting back to previous MTU %d\n", + __func__, mtu, err, prev_mtu); + + mlx5e_set_mtu(mdev, params, prev_mtu); + return err; + } + } + params->sw_mtu = mtu; return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index d873923..cb16508 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -3734,6 +3734,13 @@ static int mlx5_fs_mode_validate(struct devlink *devlink, u32 id, char *value = val.vstr; u8 eswitch_mode; + eswitch_mode = mlx5_eswitch_mode(dev); + if (eswitch_mode == MLX5_ESWITCH_OFFLOADS) { + NL_SET_ERR_MSG_FMT_MOD(extack, + "Changing fs mode is not supported when eswitch offloads enabled."); + return -EOPNOTSUPP; + } + if (!strcmp(value, "dmfs")) return 0; @@ -3759,14 +3766,6 @@ static int mlx5_fs_mode_validate(struct devlink *devlink, u32 id, return -EINVAL; } - eswitch_mode = mlx5_eswitch_mode(dev); - if (eswitch_mode == MLX5_ESWITCH_OFFLOADS) { - NL_SET_ERR_MSG_FMT_MOD(extack, - "Moving to %s is not supported when eswitch offloads enabled.", - value); - return -EOPNOTSUPP; - } - return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c index 69933ad..2299513 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c @@ -6,13 +6,15 @@ #include "fw_reset.h" #include "diag/fw_tracer.h" #include "lib/tout.h" +#include "sf/sf.h" enum { MLX5_FW_RESET_FLAGS_RESET_REQUESTED, MLX5_FW_RESET_FLAGS_NACK_RESET_REQUEST, MLX5_FW_RESET_FLAGS_PENDING_COMP, MLX5_FW_RESET_FLAGS_DROP_NEW_REQUESTS, - MLX5_FW_RESET_FLAGS_RELOAD_REQUIRED + MLX5_FW_RESET_FLAGS_RELOAD_REQUIRED, + MLX5_FW_RESET_FLAGS_UNLOAD_EVENT, }; struct mlx5_fw_reset { @@ -219,7 +221,7 @@ int mlx5_fw_reset_set_live_patch(struct mlx5_core_dev *dev) return mlx5_reg_mfrl_set(dev, MLX5_MFRL_REG_RESET_LEVEL0, 0, 0, false); } -static void mlx5_fw_reset_complete_reload(struct mlx5_core_dev *dev, bool unloaded) +static void mlx5_fw_reset_complete_reload(struct mlx5_core_dev *dev) { struct mlx5_fw_reset *fw_reset = dev->priv.fw_reset; struct devlink *devlink = priv_to_devlink(dev); @@ -228,8 +230,7 @@ static void mlx5_fw_reset_complete_reload(struct mlx5_core_dev *dev, bool unload if (test_bit(MLX5_FW_RESET_FLAGS_PENDING_COMP, &fw_reset->reset_flags)) { complete(&fw_reset->done); } else { - if (!unloaded) - mlx5_unload_one(dev, false); + mlx5_sync_reset_unload_flow(dev, false); if (mlx5_health_wait_pci_up(dev)) mlx5_core_err(dev, "reset reload flow aborted, PCI reads still not working\n"); else @@ -272,7 +273,7 @@ static void mlx5_sync_reset_reload_work(struct work_struct *work) mlx5_sync_reset_clear_reset_requested(dev, false); mlx5_enter_error_state(dev, true); - mlx5_fw_reset_complete_reload(dev, false); + mlx5_fw_reset_complete_reload(dev); } #define MLX5_RESET_POLL_INTERVAL (HZ / 10) @@ -428,6 +429,11 @@ static bool mlx5_is_reset_now_capable(struct mlx5_core_dev *dev, return false; } + if (!mlx5_core_is_ecpf(dev) && !mlx5_sf_table_empty(dev)) { + mlx5_core_warn(dev, "SFs should be removed before reset\n"); + return false; + } + #if IS_ENABLED(CONFIG_HOTPLUG_PCI_PCIE) if (reset_method != MLX5_MFRL_REG_PCI_RESET_METHOD_HOT_RESET) { err = mlx5_check_hotplug_interrupt(dev, bridge); @@ -586,6 +592,65 @@ static int mlx5_sync_pci_reset(struct mlx5_core_dev *dev, u8 reset_method) return err; } +void mlx5_sync_reset_unload_flow(struct mlx5_core_dev *dev, bool locked) +{ + struct mlx5_fw_reset *fw_reset = dev->priv.fw_reset; + unsigned long timeout; + int poll_freq = 20; + bool reset_action; + u8 rst_state; + int err; + + if (locked) + mlx5_unload_one_devl_locked(dev, false); + else + mlx5_unload_one(dev, false); + + if (!test_bit(MLX5_FW_RESET_FLAGS_UNLOAD_EVENT, &fw_reset->reset_flags)) + return; + + mlx5_set_fw_rst_ack(dev); + mlx5_core_warn(dev, "Sync Reset Unload done, device reset expected\n"); + + reset_action = false; + timeout = jiffies + msecs_to_jiffies(mlx5_tout_ms(dev, RESET_UNLOAD)); + do { + rst_state = mlx5_get_fw_rst_state(dev); + if (rst_state == MLX5_FW_RST_STATE_TOGGLE_REQ || + rst_state == MLX5_FW_RST_STATE_IDLE) { + reset_action = true; + break; + } + if (rst_state == MLX5_FW_RST_STATE_DROP_MODE) { + mlx5_core_info(dev, "Sync Reset Drop mode ack\n"); + mlx5_set_fw_rst_ack(dev); + poll_freq = 1000; + } + msleep(poll_freq); + } while (!time_after(jiffies, timeout)); + + if (!reset_action) { + mlx5_core_err(dev, "Got timeout waiting for sync reset action, state = %u\n", + rst_state); + fw_reset->ret = -ETIMEDOUT; + goto done; + } + + mlx5_core_warn(dev, "Sync Reset, got reset action. rst_state = %u\n", + rst_state); + if (rst_state == MLX5_FW_RST_STATE_TOGGLE_REQ) { + err = mlx5_sync_pci_reset(dev, fw_reset->reset_method); + if (err) { + mlx5_core_warn(dev, "mlx5_sync_pci_reset failed, err %d\n", + err); + fw_reset->ret = err; + } + } + +done: + clear_bit(MLX5_FW_RESET_FLAGS_UNLOAD_EVENT, &fw_reset->reset_flags); +} + static void mlx5_sync_reset_now_event(struct work_struct *work) { struct mlx5_fw_reset *fw_reset = container_of(work, struct mlx5_fw_reset, @@ -613,17 +678,13 @@ static void mlx5_sync_reset_now_event(struct work_struct *work) mlx5_enter_error_state(dev, true); done: fw_reset->ret = err; - mlx5_fw_reset_complete_reload(dev, false); + mlx5_fw_reset_complete_reload(dev); } static void mlx5_sync_reset_unload_event(struct work_struct *work) { struct mlx5_fw_reset *fw_reset; struct mlx5_core_dev *dev; - unsigned long timeout; - int poll_freq = 20; - bool reset_action; - u8 rst_state; int err; fw_reset = container_of(work, struct mlx5_fw_reset, reset_unload_work); @@ -632,6 +693,7 @@ static void mlx5_sync_reset_unload_event(struct work_struct *work) if (mlx5_sync_reset_clear_reset_requested(dev, false)) return; + set_bit(MLX5_FW_RESET_FLAGS_UNLOAD_EVENT, &fw_reset->reset_flags); mlx5_core_warn(dev, "Sync Reset Unload. Function is forced down.\n"); err = mlx5_cmd_fast_teardown_hca(dev); @@ -640,49 +702,7 @@ static void mlx5_sync_reset_unload_event(struct work_struct *work) else mlx5_enter_error_state(dev, true); - if (test_bit(MLX5_FW_RESET_FLAGS_PENDING_COMP, &fw_reset->reset_flags)) - mlx5_unload_one_devl_locked(dev, false); - else - mlx5_unload_one(dev, false); - - mlx5_set_fw_rst_ack(dev); - mlx5_core_warn(dev, "Sync Reset Unload done, device reset expected\n"); - - reset_action = false; - timeout = jiffies + msecs_to_jiffies(mlx5_tout_ms(dev, RESET_UNLOAD)); - do { - rst_state = mlx5_get_fw_rst_state(dev); - if (rst_state == MLX5_FW_RST_STATE_TOGGLE_REQ || - rst_state == MLX5_FW_RST_STATE_IDLE) { - reset_action = true; - break; - } - if (rst_state == MLX5_FW_RST_STATE_DROP_MODE) { - mlx5_core_info(dev, "Sync Reset Drop mode ack\n"); - mlx5_set_fw_rst_ack(dev); - poll_freq = 1000; - } - msleep(poll_freq); - } while (!time_after(jiffies, timeout)); - - if (!reset_action) { - mlx5_core_err(dev, "Got timeout waiting for sync reset action, state = %u\n", - rst_state); - fw_reset->ret = -ETIMEDOUT; - goto done; - } - - mlx5_core_warn(dev, "Sync Reset, got reset action. rst_state = %u\n", rst_state); - if (rst_state == MLX5_FW_RST_STATE_TOGGLE_REQ) { - err = mlx5_sync_pci_reset(dev, fw_reset->reset_method); - if (err) { - mlx5_core_warn(dev, "mlx5_sync_pci_reset failed, err %d\n", err); - fw_reset->ret = err; - } - } - -done: - mlx5_fw_reset_complete_reload(dev, true); + mlx5_fw_reset_complete_reload(dev); } static void mlx5_sync_reset_abort_event(struct work_struct *work) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.h b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.h index ea527d0..d5b2852 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.h @@ -12,6 +12,7 @@ int mlx5_fw_reset_set_reset_sync(struct mlx5_core_dev *dev, u8 reset_type_sel, int mlx5_fw_reset_set_live_patch(struct mlx5_core_dev *dev); int mlx5_fw_reset_wait_reset_done(struct mlx5_core_dev *dev); +void mlx5_sync_reset_unload_flow(struct mlx5_core_dev *dev, bool locked); int mlx5_fw_reset_verify_fw_complete(struct mlx5_core_dev *dev, struct netlink_ext_ack *extack); void mlx5_fw_reset_events_start(struct mlx5_core_dev *dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c index 0864ba6..3304f25 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c @@ -518,3 +518,13 @@ void mlx5_sf_table_cleanup(struct mlx5_core_dev *dev) WARN_ON(!xa_empty(&table->function_ids)); kfree(table); } + +bool mlx5_sf_table_empty(const struct mlx5_core_dev *dev) +{ + struct mlx5_sf_table *table = dev->priv.sf_table; + + if (!table) + return true; + + return xa_empty(&table->function_ids); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/sf.h b/drivers/net/ethernet/mellanox/mlx5/core/sf/sf.h index 860f9dd..89559a3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/sf.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/sf.h @@ -17,6 +17,7 @@ void mlx5_sf_hw_table_destroy(struct mlx5_core_dev *dev); int mlx5_sf_table_init(struct mlx5_core_dev *dev); void mlx5_sf_table_cleanup(struct mlx5_core_dev *dev); +bool mlx5_sf_table_empty(const struct mlx5_core_dev *dev); int mlx5_devlink_sf_port_new(struct devlink *devlink, const struct devlink_port_new_attrs *add_attr, @@ -61,6 +62,11 @@ static inline void mlx5_sf_table_cleanup(struct mlx5_core_dev *dev) { } +static inline bool mlx5_sf_table_empty(const struct mlx5_core_dev *dev) +{ + return true; +} + #endif #endif diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/action.c index 3968043..6b36a4a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/action.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/action.c @@ -117,7 +117,7 @@ static int hws_action_get_shared_stc_nic(struct mlx5hws_context *ctx, mlx5hws_err(ctx, "No such stc_type: %d\n", stc_type); pr_warn("HWS: Invalid stc_type: %d\n", stc_type); ret = -EINVAL; - goto unlock_and_out; + goto free_shared_stc; } ret = mlx5hws_action_alloc_single_stc(ctx, &stc_attr, tbl_type, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/pat_arg.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/pat_arg.c index 51e4c55..d56271a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/pat_arg.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/pat_arg.c @@ -279,7 +279,7 @@ int mlx5hws_pat_get_pattern(struct mlx5hws_context *ctx, return ret; clean_pattern: - mlx5hws_cmd_header_modify_pattern_destroy(ctx->mdev, *pattern_id); + mlx5hws_cmd_header_modify_pattern_destroy(ctx->mdev, ptrn_id); out_unlock: mutex_unlock(&ctx->pattern_cache->lock); return ret; @@ -527,7 +527,6 @@ int mlx5hws_pat_calc_nop(__be64 *pattern, size_t num_actions, u32 *nop_locations, __be64 *new_pat) { u16 prev_src_field = INVALID_FIELD, prev_dst_field = INVALID_FIELD; - u16 src_field, dst_field; u8 action_type; bool dependent; size_t i, j; @@ -539,6 +538,9 @@ int mlx5hws_pat_calc_nop(__be64 *pattern, size_t num_actions, return 0; for (i = 0, j = 0; i < num_actions; i++, j++) { + u16 src_field = INVALID_FIELD; + u16 dst_field = INVALID_FIELD; + if (j >= max_actions) return -EINVAL; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/pool.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/pool.c index 7e37d6e..7b5071c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/pool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/pool.c @@ -124,6 +124,7 @@ static int hws_pool_buddy_init(struct mlx5hws_pool *pool) mlx5hws_err(pool->ctx, "Failed to create resource type: %d size %zu\n", pool->type, pool->alloc_log_sz); mlx5hws_buddy_cleanup(buddy); + kfree(buddy); return -ENOMEM; } diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c index e67e994..4058155 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c @@ -52,6 +52,8 @@ int __fbnic_open(struct fbnic_net *fbn) fbnic_bmc_rpc_init(fbd); fbnic_rss_reinit(fbd, fbn); + phylink_resume(fbn->phylink); + return 0; time_stop: fbnic_time_stop(fbn); @@ -84,6 +86,8 @@ static int fbnic_stop(struct net_device *netdev) { struct fbnic_net *fbn = netdev_priv(netdev); + phylink_suspend(fbn->phylink, fbnic_bmc_present(fbn->fbd)); + fbnic_down(fbn); fbnic_pcs_free_irq(fbn->fbd); diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c index b70e4ca..28e23e3 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c @@ -118,14 +118,12 @@ static void fbnic_service_task_start(struct fbnic_net *fbn) struct fbnic_dev *fbd = fbn->fbd; schedule_delayed_work(&fbd->service_task, HZ); - phylink_resume(fbn->phylink); } static void fbnic_service_task_stop(struct fbnic_net *fbn) { struct fbnic_dev *fbd = fbn->fbd; - phylink_suspend(fbn->phylink, fbnic_bmc_present(fbd)); cancel_delayed_work(&fbd->service_task); } @@ -443,11 +441,10 @@ static int __fbnic_pm_resume(struct device *dev) /* Re-enable mailbox */ err = fbnic_fw_request_mbx(fbd); + devl_unlock(priv_to_devlink(fbd)); if (err) goto err_free_irqs; - devl_unlock(priv_to_devlink(fbd)); - /* Only send log history if log buffer is empty to prevent duplicate * log entries. */ @@ -464,20 +461,20 @@ static int __fbnic_pm_resume(struct device *dev) rtnl_lock(); - if (netif_running(netdev)) { + if (netif_running(netdev)) err = __fbnic_open(fbn); - if (err) - goto err_free_mbx; - } rtnl_unlock(); + if (err) + goto err_free_mbx; return 0; err_free_mbx: fbnic_fw_log_disable(fbd); - rtnl_unlock(); + devl_lock(priv_to_devlink(fbd)); fbnic_fw_free_mbx(fbd); + devl_unlock(priv_to_devlink(fbd)); err_free_irqs: fbnic_free_irqs(fbd); err_invalidate_uc_addr: diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c index 6cadf8d..00e929b 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c @@ -49,6 +49,14 @@ static void dwxgmac2_core_init(struct mac_device_info *hw, writel(XGMAC_INT_DEFAULT_EN, ioaddr + XGMAC_INT_EN); } +static void dwxgmac2_update_caps(struct stmmac_priv *priv) +{ + if (!priv->dma_cap.mbps_10_100) + priv->hw->link.caps &= ~(MAC_10 | MAC_100); + else if (!priv->dma_cap.half_duplex) + priv->hw->link.caps &= ~(MAC_10HD | MAC_100HD); +} + static void dwxgmac2_set_mac(void __iomem *ioaddr, bool enable) { u32 tx = readl(ioaddr + XGMAC_TX_CONFIG); @@ -1424,6 +1432,7 @@ static void dwxgmac2_set_arp_offload(struct mac_device_info *hw, bool en, const struct stmmac_ops dwxgmac210_ops = { .core_init = dwxgmac2_core_init, + .update_caps = dwxgmac2_update_caps, .set_mac = dwxgmac2_set_mac, .rx_ipc = dwxgmac2_rx_ipc, .rx_queue_enable = dwxgmac2_rx_queue_enable, @@ -1532,8 +1541,8 @@ int dwxgmac2_setup(struct stmmac_priv *priv) mac->mcast_bits_log2 = ilog2(mac->multicast_filter_bins); mac->link.caps = MAC_ASYM_PAUSE | MAC_SYM_PAUSE | - MAC_1000FD | MAC_2500FD | MAC_5000FD | - MAC_10000FD; + MAC_10 | MAC_100 | MAC_1000FD | + MAC_2500FD | MAC_5000FD | MAC_10000FD; mac->link.duplex = 0; mac->link.speed10 = XGMAC_CONFIG_SS_10_MII; mac->link.speed100 = XGMAC_CONFIG_SS_100_MII; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c index 5dcc95b..4d6bb99 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c @@ -203,10 +203,6 @@ static void dwxgmac2_dma_rx_mode(struct stmmac_priv *priv, void __iomem *ioaddr, } writel(value, ioaddr + XGMAC_MTL_RXQ_OPMODE(channel)); - - /* Enable MTL RX overflow */ - value = readl(ioaddr + XGMAC_MTL_QINTEN(channel)); - writel(value | XGMAC_RXOIE, ioaddr + XGMAC_MTL_QINTEN(channel)); } static void dwxgmac2_dma_tx_mode(struct stmmac_priv *priv, void __iomem *ioaddr, @@ -386,8 +382,11 @@ static int dwxgmac2_dma_interrupt(struct stmmac_priv *priv, static int dwxgmac2_get_hw_feature(void __iomem *ioaddr, struct dma_features *dma_cap) { + struct stmmac_priv *priv; u32 hw_cap; + priv = container_of(dma_cap, struct stmmac_priv, dma_cap); + /* MAC HW feature 0 */ hw_cap = readl(ioaddr + XGMAC_HW_FEATURE0); dma_cap->edma = (hw_cap & XGMAC_HWFEAT_EDMA) >> 31; @@ -410,6 +409,8 @@ static int dwxgmac2_get_hw_feature(void __iomem *ioaddr, dma_cap->vlhash = (hw_cap & XGMAC_HWFEAT_VLHASH) >> 4; dma_cap->half_duplex = (hw_cap & XGMAC_HWFEAT_HDSEL) >> 3; dma_cap->mbps_1000 = (hw_cap & XGMAC_HWFEAT_GMIISEL) >> 1; + if (dma_cap->mbps_1000 && priv->synopsys_id >= DWXGMAC_CORE_2_20) + dma_cap->mbps_10_100 = 1; /* MAC HW feature 1 */ hw_cap = readl(ioaddr + XGMAC_HW_FEATURE1); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index f1abf42..7b16d12 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -2584,6 +2584,7 @@ static bool stmmac_xdp_xmit_zc(struct stmmac_priv *priv, u32 queue, u32 budget) struct netdev_queue *nq = netdev_get_tx_queue(priv->dev, queue); struct stmmac_tx_queue *tx_q = &priv->dma_conf.tx_queue[queue]; struct stmmac_txq_stats *txq_stats = &priv->xstats.txq_stats[queue]; + bool csum = !priv->plat->tx_queues_cfg[queue].coe_unsupported; struct xsk_buff_pool *pool = tx_q->xsk_pool; unsigned int entry = tx_q->cur_tx; struct dma_desc *tx_desc = NULL; @@ -2671,7 +2672,7 @@ static bool stmmac_xdp_xmit_zc(struct stmmac_priv *priv, u32 queue, u32 budget) } stmmac_prepare_tx_desc(priv, tx_desc, 1, xdp_desc.len, - true, priv->mode, true, true, + csum, priv->mode, true, true, xdp_desc.len); stmmac_enable_dma_transmission(priv, priv->ioaddr, queue); @@ -4983,6 +4984,7 @@ static int stmmac_xdp_xmit_xdpf(struct stmmac_priv *priv, int queue, { struct stmmac_txq_stats *txq_stats = &priv->xstats.txq_stats[queue]; struct stmmac_tx_queue *tx_q = &priv->dma_conf.tx_queue[queue]; + bool csum = !priv->plat->tx_queues_cfg[queue].coe_unsupported; unsigned int entry = tx_q->cur_tx; struct dma_desc *tx_desc; dma_addr_t dma_addr; @@ -5034,7 +5036,7 @@ static int stmmac_xdp_xmit_xdpf(struct stmmac_priv *priv, int queue, stmmac_set_desc_addr(priv, tx_desc, dma_addr); stmmac_prepare_tx_desc(priv, tx_desc, 1, xdpf->len, - true, priv->mode, true, true, + csum, priv->mode, true, true, xdpf->len); tx_q->tx_count_frames++; diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c index 7201046..60a4629 100644 --- a/drivers/net/hyperv/netvsc.c +++ b/drivers/net/hyperv/netvsc.c @@ -1812,6 +1812,11 @@ struct netvsc_device *netvsc_device_add(struct hv_device *device, /* Enable NAPI handler before init callbacks */ netif_napi_add(ndev, &net_device->chan_table[0].napi, netvsc_poll); + napi_enable(&net_device->chan_table[0].napi); + netif_queue_set_napi(ndev, 0, NETDEV_QUEUE_TYPE_RX, + &net_device->chan_table[0].napi); + netif_queue_set_napi(ndev, 0, NETDEV_QUEUE_TYPE_TX, + &net_device->chan_table[0].napi); /* Open the channel */ device->channel->next_request_id_callback = vmbus_next_request_id; @@ -1831,12 +1836,6 @@ struct netvsc_device *netvsc_device_add(struct hv_device *device, /* Channel is opened */ netdev_dbg(ndev, "hv_netvsc channel opened successfully\n"); - napi_enable(&net_device->chan_table[0].napi); - netif_queue_set_napi(ndev, 0, NETDEV_QUEUE_TYPE_RX, - &net_device->chan_table[0].napi); - netif_queue_set_napi(ndev, 0, NETDEV_QUEUE_TYPE_TX, - &net_device->chan_table[0].napi); - /* Connect with the NetVsp */ ret = netvsc_connect_vsp(device, net_device, device_info); if (ret != 0) { @@ -1854,14 +1853,14 @@ struct netvsc_device *netvsc_device_add(struct hv_device *device, close: RCU_INIT_POINTER(net_device_ctx->nvdev, NULL); - netif_queue_set_napi(ndev, 0, NETDEV_QUEUE_TYPE_TX, NULL); - netif_queue_set_napi(ndev, 0, NETDEV_QUEUE_TYPE_RX, NULL); - napi_disable(&net_device->chan_table[0].napi); /* Now, we can close the channel safely */ vmbus_close(device->channel); cleanup: + netif_queue_set_napi(ndev, 0, NETDEV_QUEUE_TYPE_TX, NULL); + netif_queue_set_napi(ndev, 0, NETDEV_QUEUE_TYPE_RX, NULL); + napi_disable(&net_device->chan_table[0].napi); netif_napi_del(&net_device->chan_table[0].napi); cleanup2: diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c index 9e73959..c35f968 100644 --- a/drivers/net/hyperv/rndis_filter.c +++ b/drivers/net/hyperv/rndis_filter.c @@ -1252,17 +1252,26 @@ static void netvsc_sc_open(struct vmbus_channel *new_sc) new_sc->rqstor_size = netvsc_rqstor_size(netvsc_ring_bytes); new_sc->max_pkt_size = NETVSC_MAX_PKT_SIZE; + /* Enable napi before opening the vmbus channel to avoid races + * as the host placing data on the host->guest ring may be left + * out if napi was not enabled. + */ + napi_enable(&nvchan->napi); + netif_queue_set_napi(ndev, chn_index, NETDEV_QUEUE_TYPE_RX, + &nvchan->napi); + netif_queue_set_napi(ndev, chn_index, NETDEV_QUEUE_TYPE_TX, + &nvchan->napi); + ret = vmbus_open(new_sc, netvsc_ring_bytes, netvsc_ring_bytes, NULL, 0, netvsc_channel_cb, nvchan); - if (ret == 0) { - napi_enable(&nvchan->napi); - netif_queue_set_napi(ndev, chn_index, NETDEV_QUEUE_TYPE_RX, - &nvchan->napi); - netif_queue_set_napi(ndev, chn_index, NETDEV_QUEUE_TYPE_TX, - &nvchan->napi); - } else { + if (ret != 0) { netdev_notice(ndev, "sub channel open failed: %d\n", ret); + netif_queue_set_napi(ndev, chn_index, NETDEV_QUEUE_TYPE_TX, + NULL); + netif_queue_set_napi(ndev, chn_index, NETDEV_QUEUE_TYPE_RX, + NULL); + napi_disable(&nvchan->napi); } if (atomic_inc_return(&nvscdev->open_chn) == nvscdev->num_chn) diff --git a/drivers/net/phy/mscc/mscc.h b/drivers/net/phy/mscc/mscc.h index 58c6d47..2bfe314 100644 --- a/drivers/net/phy/mscc/mscc.h +++ b/drivers/net/phy/mscc/mscc.h @@ -481,6 +481,7 @@ static inline void vsc8584_config_macsec_intr(struct phy_device *phydev) void vsc85xx_link_change_notify(struct phy_device *phydev); void vsc8584_config_ts_intr(struct phy_device *phydev); int vsc8584_ptp_init(struct phy_device *phydev); +void vsc8584_ptp_deinit(struct phy_device *phydev); int vsc8584_ptp_probe_once(struct phy_device *phydev); int vsc8584_ptp_probe(struct phy_device *phydev); irqreturn_t vsc8584_handle_ts_interrupt(struct phy_device *phydev); @@ -495,6 +496,9 @@ static inline int vsc8584_ptp_init(struct phy_device *phydev) { return 0; } +static inline void vsc8584_ptp_deinit(struct phy_device *phydev) +{ +} static inline int vsc8584_ptp_probe_once(struct phy_device *phydev) { return 0; diff --git a/drivers/net/phy/mscc/mscc_main.c b/drivers/net/phy/mscc/mscc_main.c index f1c9ce3..24c7590 100644 --- a/drivers/net/phy/mscc/mscc_main.c +++ b/drivers/net/phy/mscc/mscc_main.c @@ -2337,9 +2337,7 @@ static int vsc85xx_probe(struct phy_device *phydev) static void vsc85xx_remove(struct phy_device *phydev) { - struct vsc8531_private *priv = phydev->priv; - - skb_queue_purge(&priv->rx_skbs_list); + vsc8584_ptp_deinit(phydev); } /* Microsemi VSC85xx PHYs */ diff --git a/drivers/net/phy/mscc/mscc_ptp.c b/drivers/net/phy/mscc/mscc_ptp.c index de6c731..7284732 100644 --- a/drivers/net/phy/mscc/mscc_ptp.c +++ b/drivers/net/phy/mscc/mscc_ptp.c @@ -1298,7 +1298,6 @@ static void vsc8584_set_input_clk_configured(struct phy_device *phydev) static int __vsc8584_init_ptp(struct phy_device *phydev) { - struct vsc8531_private *vsc8531 = phydev->priv; static const u32 ltc_seq_e[] = { 0, 400000, 0, 0, 0 }; static const u8 ltc_seq_a[] = { 8, 6, 5, 4, 2 }; u32 val; @@ -1515,17 +1514,7 @@ static int __vsc8584_init_ptp(struct phy_device *phydev) vsc85xx_ts_eth_cmp1_sig(phydev); - vsc8531->mii_ts.rxtstamp = vsc85xx_rxtstamp; - vsc8531->mii_ts.txtstamp = vsc85xx_txtstamp; - vsc8531->mii_ts.hwtstamp = vsc85xx_hwtstamp; - vsc8531->mii_ts.ts_info = vsc85xx_ts_info; - phydev->mii_ts = &vsc8531->mii_ts; - - memcpy(&vsc8531->ptp->caps, &vsc85xx_clk_caps, sizeof(vsc85xx_clk_caps)); - - vsc8531->ptp->ptp_clock = ptp_clock_register(&vsc8531->ptp->caps, - &phydev->mdio.dev); - return PTR_ERR_OR_ZERO(vsc8531->ptp->ptp_clock); + return 0; } void vsc8584_config_ts_intr(struct phy_device *phydev) @@ -1552,6 +1541,16 @@ int vsc8584_ptp_init(struct phy_device *phydev) return 0; } +void vsc8584_ptp_deinit(struct phy_device *phydev) +{ + struct vsc8531_private *vsc8531 = phydev->priv; + + if (vsc8531->ptp->ptp_clock) { + ptp_clock_unregister(vsc8531->ptp->ptp_clock); + skb_queue_purge(&vsc8531->rx_skbs_list); + } +} + irqreturn_t vsc8584_handle_ts_interrupt(struct phy_device *phydev) { struct vsc8531_private *priv = phydev->priv; @@ -1612,7 +1611,16 @@ int vsc8584_ptp_probe(struct phy_device *phydev) vsc8531->ptp->phydev = phydev; - return 0; + vsc8531->mii_ts.rxtstamp = vsc85xx_rxtstamp; + vsc8531->mii_ts.txtstamp = vsc85xx_txtstamp; + vsc8531->mii_ts.hwtstamp = vsc85xx_hwtstamp; + vsc8531->mii_ts.ts_info = vsc85xx_ts_info; + phydev->mii_ts = &vsc8531->mii_ts; + + memcpy(&vsc8531->ptp->caps, &vsc85xx_clk_caps, sizeof(vsc85xx_clk_caps)); + vsc8531->ptp->ptp_clock = ptp_clock_register(&vsc8531->ptp->caps, + &phydev->mdio.dev); + return PTR_ERR_OR_ZERO(vsc8531->ptp->ptp_clock); } int vsc8584_ptp_probe_once(struct phy_device *phydev) diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c index e56901b..11352d8 100644 --- a/drivers/net/usb/qmi_wwan.c +++ b/drivers/net/usb/qmi_wwan.c @@ -1355,6 +1355,9 @@ static const struct usb_device_id products[] = { {QMI_FIXED_INTF(0x2357, 0x0201, 4)}, /* TP-LINK HSUPA Modem MA180 */ {QMI_FIXED_INTF(0x2357, 0x9000, 4)}, /* TP-LINK MA260 */ {QMI_QUIRK_SET_DTR(0x1bc7, 0x1031, 3)}, /* Telit LE910C1-EUX */ + {QMI_QUIRK_SET_DTR(0x1bc7, 0x1034, 2)}, /* Telit LE910C4-WWX */ + {QMI_QUIRK_SET_DTR(0x1bc7, 0x1037, 4)}, /* Telit LE910C4-WWX */ + {QMI_QUIRK_SET_DTR(0x1bc7, 0x1038, 3)}, /* Telit LE910C4-WWX */ {QMI_QUIRK_SET_DTR(0x1bc7, 0x103a, 0)}, /* Telit LE910C4-WWX */ {QMI_QUIRK_SET_DTR(0x1bc7, 0x1040, 2)}, /* Telit LE922A */ {QMI_QUIRK_SET_DTR(0x1bc7, 0x1050, 2)}, /* Telit FN980 */ diff --git a/include/linux/atmdev.h b/include/linux/atmdev.h index 45f2f27..70807c6 100644 --- a/include/linux/atmdev.h +++ b/include/linux/atmdev.h @@ -185,6 +185,7 @@ struct atmdev_ops { /* only send is required */ int (*compat_ioctl)(struct atm_dev *dev,unsigned int cmd, void __user *arg); #endif + int (*pre_send)(struct atm_vcc *vcc, struct sk_buff *skb); int (*send)(struct atm_vcc *vcc,struct sk_buff *skb); int (*send_bh)(struct atm_vcc *vcc, struct sk_buff *skb); int (*send_oam)(struct atm_vcc *vcc,void *cell,int flags); diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 14b923d..fa63365 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -4172,6 +4172,8 @@ int skb_copy_and_crc32c_datagram_iter(const struct sk_buff *skb, int offset, struct iov_iter *to, int len, u32 *crcp); int skb_copy_datagram_from_iter(struct sk_buff *skb, int offset, struct iov_iter *from, int len); +int skb_copy_datagram_from_iter_full(struct sk_buff *skb, int offset, + struct iov_iter *from, int len); int zerocopy_sg_from_iter(struct sk_buff *skb, struct iov_iter *frm); void skb_free_datagram(struct sock *sk, struct sk_buff *skb); int skb_kill_datagram(struct sock *sk, struct sk_buff *skb, unsigned int flags); diff --git a/include/net/bluetooth/hci_sync.h b/include/net/bluetooth/hci_sync.h index 5224f57..e352a4e 100644 --- a/include/net/bluetooth/hci_sync.h +++ b/include/net/bluetooth/hci_sync.h @@ -93,7 +93,7 @@ int hci_update_class_sync(struct hci_dev *hdev); int hci_update_eir_sync(struct hci_dev *hdev); int hci_update_class_sync(struct hci_dev *hdev); -int hci_update_name_sync(struct hci_dev *hdev); +int hci_update_name_sync(struct hci_dev *hdev, const u8 *name); int hci_write_ssp_mode_sync(struct hci_dev *hdev, u8 mode); int hci_get_random_address(struct hci_dev *hdev, bool require_privacy, diff --git a/include/net/rose.h b/include/net/rose.h index 23267b4..2b5491b 100644 --- a/include/net/rose.h +++ b/include/net/rose.h @@ -8,6 +8,7 @@ #ifndef _ROSE_H #define _ROSE_H +#include #include #include #include @@ -96,7 +97,7 @@ struct rose_neigh { ax25_cb *ax25; struct net_device *dev; unsigned short count; - unsigned short use; + refcount_t use; unsigned int number; char restarted; char dce_mode; @@ -151,6 +152,21 @@ struct rose_sock { #define rose_sk(sk) ((struct rose_sock *)(sk)) +static inline void rose_neigh_hold(struct rose_neigh *rose_neigh) +{ + refcount_inc(&rose_neigh->use); +} + +static inline void rose_neigh_put(struct rose_neigh *rose_neigh) +{ + if (refcount_dec_and_test(&rose_neigh->use)) { + if (rose_neigh->ax25) + ax25_cb_put(rose_neigh->ax25); + kfree(rose_neigh->digipeat); + kfree(rose_neigh); + } +} + /* af_rose.c */ extern ax25_address rose_callsign; extern int sysctl_rose_restart_request_timeout; diff --git a/net/atm/common.c b/net/atm/common.c index d7f7976..881c7f2 100644 --- a/net/atm/common.c +++ b/net/atm/common.c @@ -635,18 +635,27 @@ int vcc_sendmsg(struct socket *sock, struct msghdr *m, size_t size) skb->dev = NULL; /* for paths shared with net_device interfaces */ if (!copy_from_iter_full(skb_put(skb, size), size, &m->msg_iter)) { - atm_return_tx(vcc, skb); - kfree_skb(skb); error = -EFAULT; - goto out; + goto free_skb; } if (eff != size) memset(skb->data + size, 0, eff-size); + + if (vcc->dev->ops->pre_send) { + error = vcc->dev->ops->pre_send(vcc, skb); + if (error) + goto free_skb; + } + error = vcc->dev->ops->send(vcc, skb); error = error ? error : size; out: release_sock(sk); return error; +free_skb: + atm_return_tx(vcc, skb); + kfree_skb(skb); + goto out; } __poll_t vcc_poll(struct file *file, struct socket *sock, poll_table *wait) diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 7a87929..e524bb5 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -149,8 +149,6 @@ static void hci_conn_cleanup(struct hci_conn *conn) hci_chan_list_flush(conn); - hci_conn_hash_del(hdev, conn); - if (HCI_CONN_HANDLE_UNSET(conn->handle)) ida_free(&hdev->unset_handle_ida, conn->handle); @@ -1152,28 +1150,54 @@ void hci_conn_del(struct hci_conn *conn) disable_delayed_work_sync(&conn->auto_accept_work); disable_delayed_work_sync(&conn->idle_work); - if (conn->type == ACL_LINK) { - /* Unacked frames */ - hdev->acl_cnt += conn->sent; - } else if (conn->type == LE_LINK) { - cancel_delayed_work(&conn->le_conn_timeout); + /* Remove the connection from the list so unacked logic can detect when + * a certain pool is not being utilized. + */ + hci_conn_hash_del(hdev, conn); - if (hdev->le_pkts) - hdev->le_cnt += conn->sent; + /* Handle unacked frames: + * + * - In case there are no connection, or if restoring the buffers + * considered in transist would overflow, restore all buffers to the + * pool. + * - Otherwise restore just the buffers considered in transit for the + * hci_conn + */ + switch (conn->type) { + case ACL_LINK: + if (!hci_conn_num(hdev, ACL_LINK) || + hdev->acl_cnt + conn->sent > hdev->acl_pkts) + hdev->acl_cnt = hdev->acl_pkts; else hdev->acl_cnt += conn->sent; - } else { - /* Unacked ISO frames */ - if (conn->type == CIS_LINK || - conn->type == BIS_LINK || - conn->type == PA_LINK) { - if (hdev->iso_pkts) - hdev->iso_cnt += conn->sent; - else if (hdev->le_pkts) + break; + case LE_LINK: + cancel_delayed_work(&conn->le_conn_timeout); + + if (hdev->le_pkts) { + if (!hci_conn_num(hdev, LE_LINK) || + hdev->le_cnt + conn->sent > hdev->le_pkts) + hdev->le_cnt = hdev->le_pkts; + else hdev->le_cnt += conn->sent; + } else { + if ((!hci_conn_num(hdev, LE_LINK) && + !hci_conn_num(hdev, ACL_LINK)) || + hdev->acl_cnt + conn->sent > hdev->acl_pkts) + hdev->acl_cnt = hdev->acl_pkts; else hdev->acl_cnt += conn->sent; } + break; + case CIS_LINK: + case BIS_LINK: + case PA_LINK: + if (!hci_iso_count(hdev) || + hdev->iso_cnt + conn->sent > hdev->iso_pkts) + hdev->iso_cnt = hdev->iso_pkts; + else + hdev->iso_cnt += conn->sent; + break; } skb_queue_purge(&conn->data_q); diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index fe7cdd6..7a21748 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -2703,7 +2703,7 @@ static void hci_cs_disconnect(struct hci_dev *hdev, u8 status) if (!conn) goto unlock; - if (status) { + if (status && status != HCI_ERROR_UNKNOWN_CONN_ID) { mgmt_disconnect_failed(hdev, &conn->dst, conn->type, conn->dst_type, status); @@ -2718,6 +2718,12 @@ static void hci_cs_disconnect(struct hci_dev *hdev, u8 status) goto done; } + /* During suspend, mark connection as closed immediately + * since we might not receive HCI_EV_DISCONN_COMPLETE + */ + if (hdev->suspended) + conn->state = BT_CLOSED; + mgmt_conn = test_and_clear_bit(HCI_CONN_MGMT_CONNECTED, &conn->flags); if (conn->type == ACL_LINK) { @@ -4398,7 +4404,17 @@ static void hci_num_comp_pkts_evt(struct hci_dev *hdev, void *data, if (!conn) continue; - conn->sent -= count; + /* Check if there is really enough packets outstanding before + * attempting to decrease the sent counter otherwise it could + * underflow.. + */ + if (conn->sent >= count) { + conn->sent -= count; + } else { + bt_dev_warn(hdev, "hcon %p sent %u < count %u", + conn, conn->sent, count); + conn->sent = 0; + } for (i = 0; i < count; ++i) hci_conn_tx_dequeue(conn); @@ -7008,6 +7024,7 @@ static void hci_le_big_sync_lost_evt(struct hci_dev *hdev, void *data, { struct hci_evt_le_big_sync_lost *ev = data; struct hci_conn *bis, *conn; + bool mgmt_conn; bt_dev_dbg(hdev, "big handle 0x%2.2x", ev->handle); @@ -7026,6 +7043,10 @@ static void hci_le_big_sync_lost_evt(struct hci_dev *hdev, void *data, while ((bis = hci_conn_hash_lookup_big_state(hdev, ev->handle, BT_CONNECTED, HCI_ROLE_SLAVE))) { + mgmt_conn = test_and_clear_bit(HCI_CONN_MGMT_CONNECTED, &bis->flags); + mgmt_device_disconnected(hdev, &bis->dst, bis->type, bis->dst_type, + ev->reason, mgmt_conn); + clear_bit(HCI_CONN_BIG_SYNC, &bis->flags); hci_disconn_cfm(bis, ev->reason); hci_conn_del(bis); diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 31d72b9..b6f888d 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -3481,13 +3481,13 @@ int hci_update_scan_sync(struct hci_dev *hdev) return hci_write_scan_enable_sync(hdev, scan); } -int hci_update_name_sync(struct hci_dev *hdev) +int hci_update_name_sync(struct hci_dev *hdev, const u8 *name) { struct hci_cp_write_local_name cp; memset(&cp, 0, sizeof(cp)); - memcpy(cp.name, hdev->dev_name, sizeof(cp.name)); + memcpy(cp.name, name, sizeof(cp.name)); return __hci_cmd_sync_status(hdev, HCI_OP_WRITE_LOCAL_NAME, sizeof(cp), &cp, @@ -3540,7 +3540,7 @@ int hci_powered_update_sync(struct hci_dev *hdev) hci_write_fast_connectable_sync(hdev, false); hci_update_scan_sync(hdev); hci_update_class_sync(hdev); - hci_update_name_sync(hdev); + hci_update_name_sync(hdev, hdev->dev_name); hci_update_eir_sync(hdev); } diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 3166f5f..50634ef 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -3892,8 +3892,11 @@ static void set_name_complete(struct hci_dev *hdev, void *data, int err) static int set_name_sync(struct hci_dev *hdev, void *data) { + struct mgmt_pending_cmd *cmd = data; + struct mgmt_cp_set_local_name *cp = cmd->param; + if (lmp_bredr_capable(hdev)) { - hci_update_name_sync(hdev); + hci_update_name_sync(hdev, cp->name); hci_update_eir_sync(hdev); } @@ -9705,7 +9708,9 @@ void mgmt_device_disconnected(struct hci_dev *hdev, bdaddr_t *bdaddr, if (!mgmt_connected) return; - if (link_type != ACL_LINK && link_type != LE_LINK) + if (link_type != ACL_LINK && + link_type != LE_LINK && + link_type != BIS_LINK) return; bacpy(&ev.addr.bdaddr, bdaddr); diff --git a/net/core/datagram.c b/net/core/datagram.c index 94cc470..f474b9b 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -618,6 +618,20 @@ fault: } EXPORT_SYMBOL(skb_copy_datagram_from_iter); +int skb_copy_datagram_from_iter_full(struct sk_buff *skb, int offset, + struct iov_iter *from, int len) +{ + struct iov_iter_state state; + int ret; + + iov_iter_save_state(from, &state); + ret = skb_copy_datagram_from_iter(skb, offset, from, len); + if (ret) + iov_iter_restore(from, &state); + return ret; +} +EXPORT_SYMBOL(skb_copy_datagram_from_iter_full); + int zerocopy_fill_skb_from_iter(struct sk_buff *skb, struct iov_iter *from, size_t length) { diff --git a/net/core/page_pool.c b/net/core/page_pool.c index 343a6ca..ba70569 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -287,8 +287,10 @@ static int page_pool_init(struct page_pool *pool, } if (pool->mp_ops) { - if (!pool->dma_map || !pool->dma_sync) - return -EOPNOTSUPP; + if (!pool->dma_map || !pool->dma_sync) { + err = -EOPNOTSUPP; + goto free_ptr_ring; + } if (WARN_ON(!is_kernel_rodata((unsigned long)pool->mp_ops))) { err = -EFAULT; diff --git a/net/ipv4/route.c b/net/ipv4/route.c index f639a2a..baa43e5 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -2575,12 +2575,16 @@ static struct rtable *__mkroute_output(const struct fib_result *res, !netif_is_l3_master(dev_out)) return ERR_PTR(-EINVAL); - if (ipv4_is_lbcast(fl4->daddr)) + if (ipv4_is_lbcast(fl4->daddr)) { type = RTN_BROADCAST; - else if (ipv4_is_multicast(fl4->daddr)) + + /* reset fi to prevent gateway resolution */ + fi = NULL; + } else if (ipv4_is_multicast(fl4->daddr)) { type = RTN_MULTICAST; - else if (ipv4_is_zeronet(fl4->daddr)) + } else if (ipv4_is_zeronet(fl4->daddr)) { return ERR_PTR(-EINVAL); + } if (dev_out->flags & IFF_LOOPBACK) flags |= RTCF_LOCAL; diff --git a/net/l2tp/l2tp_ppp.c b/net/l2tp/l2tp_ppp.c index fc5c2fd..5e12e7c 100644 --- a/net/l2tp/l2tp_ppp.c +++ b/net/l2tp/l2tp_ppp.c @@ -129,22 +129,12 @@ static const struct ppp_channel_ops pppol2tp_chan_ops = { static const struct proto_ops pppol2tp_ops; -/* Retrieves the pppol2tp socket associated to a session. - * A reference is held on the returned socket, so this function must be paired - * with sock_put(). - */ +/* Retrieves the pppol2tp socket associated to a session. */ static struct sock *pppol2tp_session_get_sock(struct l2tp_session *session) { struct pppol2tp_session *ps = l2tp_session_priv(session); - struct sock *sk; - - rcu_read_lock(); - sk = rcu_dereference(ps->sk); - if (sk) - sock_hold(sk); - rcu_read_unlock(); - return sk; + return rcu_dereference(ps->sk); } /* Helpers to obtain tunnel/session contexts from sockets. @@ -206,14 +196,13 @@ end: static void pppol2tp_recv(struct l2tp_session *session, struct sk_buff *skb, int data_len) { - struct pppol2tp_session *ps = l2tp_session_priv(session); - struct sock *sk = NULL; + struct sock *sk; /* If the socket is bound, send it in to PPP's input queue. Otherwise * queue it on the session socket. */ rcu_read_lock(); - sk = rcu_dereference(ps->sk); + sk = pppol2tp_session_get_sock(session); if (!sk) goto no_sock; @@ -510,13 +499,14 @@ static void pppol2tp_show(struct seq_file *m, void *arg) struct l2tp_session *session = arg; struct sock *sk; + rcu_read_lock(); sk = pppol2tp_session_get_sock(session); if (sk) { struct pppox_sock *po = pppox_sk(sk); seq_printf(m, " interface %s\n", ppp_dev_name(&po->chan)); - sock_put(sk); } + rcu_read_unlock(); } static void pppol2tp_session_init(struct l2tp_session *session) @@ -1530,6 +1520,7 @@ static void pppol2tp_seq_session_show(struct seq_file *m, void *v) port = ntohs(inet->inet_sport); } + rcu_read_lock(); sk = pppol2tp_session_get_sock(session); if (sk) { state = sk->sk_state; @@ -1565,8 +1556,8 @@ static void pppol2tp_seq_session_show(struct seq_file *m, void *v) struct pppox_sock *po = pppox_sk(sk); seq_printf(m, " interface %s\n", ppp_dev_name(&po->chan)); - sock_put(sk); } + rcu_read_unlock(); } static int pppol2tp_seq_show(struct seq_file *m, void *v) diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c index 4e72b63..543f9e8 100644 --- a/net/rose/af_rose.c +++ b/net/rose/af_rose.c @@ -170,7 +170,7 @@ void rose_kill_by_neigh(struct rose_neigh *neigh) if (rose->neighbour == neigh) { rose_disconnect(s, ENETUNREACH, ROSE_OUT_OF_ORDER, 0); - rose->neighbour->use--; + rose_neigh_put(rose->neighbour); rose->neighbour = NULL; } } @@ -212,7 +212,7 @@ start: if (rose->device == dev) { rose_disconnect(sk, ENETUNREACH, ROSE_OUT_OF_ORDER, 0); if (rose->neighbour) - rose->neighbour->use--; + rose_neigh_put(rose->neighbour); netdev_put(rose->device, &rose->dev_tracker); rose->device = NULL; } @@ -655,7 +655,7 @@ static int rose_release(struct socket *sock) break; case ROSE_STATE_2: - rose->neighbour->use--; + rose_neigh_put(rose->neighbour); release_sock(sk); rose_disconnect(sk, 0, -1, -1); lock_sock(sk); @@ -823,6 +823,7 @@ static int rose_connect(struct socket *sock, struct sockaddr *uaddr, int addr_le rose->lci = rose_new_lci(rose->neighbour); if (!rose->lci) { err = -ENETUNREACH; + rose_neigh_put(rose->neighbour); goto out_release; } @@ -834,12 +835,14 @@ static int rose_connect(struct socket *sock, struct sockaddr *uaddr, int addr_le dev = rose_dev_first(); if (!dev) { err = -ENETUNREACH; + rose_neigh_put(rose->neighbour); goto out_release; } user = ax25_findbyuid(current_euid()); if (!user) { err = -EINVAL; + rose_neigh_put(rose->neighbour); dev_put(dev); goto out_release; } @@ -874,8 +877,6 @@ static int rose_connect(struct socket *sock, struct sockaddr *uaddr, int addr_le rose->state = ROSE_STATE_1; - rose->neighbour->use++; - rose_write_internal(sk, ROSE_CALL_REQUEST); rose_start_heartbeat(sk); rose_start_t1timer(sk); @@ -1077,7 +1078,7 @@ int rose_rx_call_request(struct sk_buff *skb, struct net_device *dev, struct ros GFP_ATOMIC); make_rose->facilities = facilities; - make_rose->neighbour->use++; + rose_neigh_hold(make_rose->neighbour); if (rose_sk(sk)->defer) { make_rose->state = ROSE_STATE_5; diff --git a/net/rose/rose_in.c b/net/rose/rose_in.c index 3e99181..0276b39 100644 --- a/net/rose/rose_in.c +++ b/net/rose/rose_in.c @@ -56,7 +56,7 @@ static int rose_state1_machine(struct sock *sk, struct sk_buff *skb, int framety case ROSE_CLEAR_REQUEST: rose_write_internal(sk, ROSE_CLEAR_CONFIRMATION); rose_disconnect(sk, ECONNREFUSED, skb->data[3], skb->data[4]); - rose->neighbour->use--; + rose_neigh_put(rose->neighbour); break; default: @@ -79,12 +79,12 @@ static int rose_state2_machine(struct sock *sk, struct sk_buff *skb, int framety case ROSE_CLEAR_REQUEST: rose_write_internal(sk, ROSE_CLEAR_CONFIRMATION); rose_disconnect(sk, 0, skb->data[3], skb->data[4]); - rose->neighbour->use--; + rose_neigh_put(rose->neighbour); break; case ROSE_CLEAR_CONFIRMATION: rose_disconnect(sk, 0, -1, -1); - rose->neighbour->use--; + rose_neigh_put(rose->neighbour); break; default: @@ -121,7 +121,7 @@ static int rose_state3_machine(struct sock *sk, struct sk_buff *skb, int framety case ROSE_CLEAR_REQUEST: rose_write_internal(sk, ROSE_CLEAR_CONFIRMATION); rose_disconnect(sk, 0, skb->data[3], skb->data[4]); - rose->neighbour->use--; + rose_neigh_put(rose->neighbour); break; case ROSE_RR: @@ -234,7 +234,7 @@ static int rose_state4_machine(struct sock *sk, struct sk_buff *skb, int framety case ROSE_CLEAR_REQUEST: rose_write_internal(sk, ROSE_CLEAR_CONFIRMATION); rose_disconnect(sk, 0, skb->data[3], skb->data[4]); - rose->neighbour->use--; + rose_neigh_put(rose->neighbour); break; default: @@ -254,7 +254,7 @@ static int rose_state5_machine(struct sock *sk, struct sk_buff *skb, int framety if (frametype == ROSE_CLEAR_REQUEST) { rose_write_internal(sk, ROSE_CLEAR_CONFIRMATION); rose_disconnect(sk, 0, skb->data[3], skb->data[4]); - rose_sk(sk)->neighbour->use--; + rose_neigh_put(rose_sk(sk)->neighbour); } return 0; diff --git a/net/rose/rose_route.c b/net/rose/rose_route.c index b72bf8a..a1e9b05 100644 --- a/net/rose/rose_route.c +++ b/net/rose/rose_route.c @@ -93,11 +93,11 @@ static int __must_check rose_add_node(struct rose_route_struct *rose_route, rose_neigh->ax25 = NULL; rose_neigh->dev = dev; rose_neigh->count = 0; - rose_neigh->use = 0; rose_neigh->dce_mode = 0; rose_neigh->loopback = 0; rose_neigh->number = rose_neigh_no++; rose_neigh->restarted = 0; + refcount_set(&rose_neigh->use, 1); skb_queue_head_init(&rose_neigh->queue); @@ -178,6 +178,7 @@ static int __must_check rose_add_node(struct rose_route_struct *rose_route, } } rose_neigh->count++; + rose_neigh_hold(rose_neigh); goto out; } @@ -187,6 +188,7 @@ static int __must_check rose_add_node(struct rose_route_struct *rose_route, rose_node->neighbour[rose_node->count] = rose_neigh; rose_node->count++; rose_neigh->count++; + rose_neigh_hold(rose_neigh); } out: @@ -234,20 +236,12 @@ static void rose_remove_neigh(struct rose_neigh *rose_neigh) if ((s = rose_neigh_list) == rose_neigh) { rose_neigh_list = rose_neigh->next; - if (rose_neigh->ax25) - ax25_cb_put(rose_neigh->ax25); - kfree(rose_neigh->digipeat); - kfree(rose_neigh); return; } while (s != NULL && s->next != NULL) { if (s->next == rose_neigh) { s->next = rose_neigh->next; - if (rose_neigh->ax25) - ax25_cb_put(rose_neigh->ax25); - kfree(rose_neigh->digipeat); - kfree(rose_neigh); return; } @@ -263,10 +257,10 @@ static void rose_remove_route(struct rose_route *rose_route) struct rose_route *s; if (rose_route->neigh1 != NULL) - rose_route->neigh1->use--; + rose_neigh_put(rose_route->neigh1); if (rose_route->neigh2 != NULL) - rose_route->neigh2->use--; + rose_neigh_put(rose_route->neigh2); if ((s = rose_route_list) == rose_route) { rose_route_list = rose_route->next; @@ -330,9 +324,12 @@ static int rose_del_node(struct rose_route_struct *rose_route, for (i = 0; i < rose_node->count; i++) { if (rose_node->neighbour[i] == rose_neigh) { rose_neigh->count--; + rose_neigh_put(rose_neigh); - if (rose_neigh->count == 0 && rose_neigh->use == 0) + if (rose_neigh->count == 0) { rose_remove_neigh(rose_neigh); + rose_neigh_put(rose_neigh); + } rose_node->count--; @@ -381,11 +378,11 @@ void rose_add_loopback_neigh(void) sn->ax25 = NULL; sn->dev = NULL; sn->count = 0; - sn->use = 0; sn->dce_mode = 1; sn->loopback = 1; sn->number = rose_neigh_no++; sn->restarted = 1; + refcount_set(&sn->use, 1); skb_queue_head_init(&sn->queue); @@ -436,6 +433,7 @@ int rose_add_loopback_node(const rose_address *address) rose_node_list = rose_node; rose_loopback_neigh->count++; + rose_neigh_hold(rose_loopback_neigh); out: spin_unlock_bh(&rose_node_list_lock); @@ -467,6 +465,7 @@ void rose_del_loopback_node(const rose_address *address) rose_remove_node(rose_node); rose_loopback_neigh->count--; + rose_neigh_put(rose_loopback_neigh); out: spin_unlock_bh(&rose_node_list_lock); @@ -506,6 +505,7 @@ void rose_rt_device_down(struct net_device *dev) memmove(&t->neighbour[i], &t->neighbour[i + 1], sizeof(t->neighbour[0]) * (t->count - i)); + rose_neigh_put(s); } if (t->count <= 0) @@ -513,6 +513,7 @@ void rose_rt_device_down(struct net_device *dev) } rose_remove_neigh(s); + rose_neigh_put(s); } spin_unlock_bh(&rose_neigh_list_lock); spin_unlock_bh(&rose_node_list_lock); @@ -548,6 +549,7 @@ static int rose_clear_routes(void) { struct rose_neigh *s, *rose_neigh; struct rose_node *t, *rose_node; + int i; spin_lock_bh(&rose_node_list_lock); spin_lock_bh(&rose_neigh_list_lock); @@ -558,17 +560,21 @@ static int rose_clear_routes(void) while (rose_node != NULL) { t = rose_node; rose_node = rose_node->next; - if (!t->loopback) + + if (!t->loopback) { + for (i = 0; i < t->count; i++) + rose_neigh_put(t->neighbour[i]); rose_remove_node(t); + } } while (rose_neigh != NULL) { s = rose_neigh; rose_neigh = rose_neigh->next; - if (s->use == 0 && !s->loopback) { - s->count = 0; + if (!s->loopback) { rose_remove_neigh(s); + rose_neigh_put(s); } } @@ -684,6 +690,7 @@ struct rose_neigh *rose_get_neigh(rose_address *addr, unsigned char *cause, for (i = 0; i < node->count; i++) { if (node->neighbour[i]->restarted) { res = node->neighbour[i]; + rose_neigh_hold(node->neighbour[i]); goto out; } } @@ -695,6 +702,7 @@ struct rose_neigh *rose_get_neigh(rose_address *addr, unsigned char *cause, for (i = 0; i < node->count; i++) { if (!rose_ftimer_running(node->neighbour[i])) { res = node->neighbour[i]; + rose_neigh_hold(node->neighbour[i]); goto out; } failed = 1; @@ -784,13 +792,13 @@ static void rose_del_route_by_neigh(struct rose_neigh *rose_neigh) } if (rose_route->neigh1 == rose_neigh) { - rose_route->neigh1->use--; + rose_neigh_put(rose_route->neigh1); rose_route->neigh1 = NULL; rose_transmit_clear_request(rose_route->neigh2, rose_route->lci2, ROSE_OUT_OF_ORDER, 0); } if (rose_route->neigh2 == rose_neigh) { - rose_route->neigh2->use--; + rose_neigh_put(rose_route->neigh2); rose_route->neigh2 = NULL; rose_transmit_clear_request(rose_route->neigh1, rose_route->lci1, ROSE_OUT_OF_ORDER, 0); } @@ -919,7 +927,7 @@ int rose_route_frame(struct sk_buff *skb, ax25_cb *ax25) rose_clear_queues(sk); rose->cause = ROSE_NETWORK_CONGESTION; rose->diagnostic = 0; - rose->neighbour->use--; + rose_neigh_put(rose->neighbour); rose->neighbour = NULL; rose->lci = 0; rose->state = ROSE_STATE_0; @@ -1044,12 +1052,12 @@ int rose_route_frame(struct sk_buff *skb, ax25_cb *ax25) if ((new_lci = rose_new_lci(new_neigh)) == 0) { rose_transmit_clear_request(rose_neigh, lci, ROSE_NETWORK_CONGESTION, 71); - goto out; + goto put_neigh; } if ((rose_route = kmalloc(sizeof(*rose_route), GFP_ATOMIC)) == NULL) { rose_transmit_clear_request(rose_neigh, lci, ROSE_NETWORK_CONGESTION, 120); - goto out; + goto put_neigh; } rose_route->lci1 = lci; @@ -1062,8 +1070,8 @@ int rose_route_frame(struct sk_buff *skb, ax25_cb *ax25) rose_route->lci2 = new_lci; rose_route->neigh2 = new_neigh; - rose_route->neigh1->use++; - rose_route->neigh2->use++; + rose_neigh_hold(rose_route->neigh1); + rose_neigh_hold(rose_route->neigh2); rose_route->next = rose_route_list; rose_route_list = rose_route; @@ -1075,6 +1083,8 @@ int rose_route_frame(struct sk_buff *skb, ax25_cb *ax25) rose_transmit_link(skb, rose_route->neigh2); res = 1; +put_neigh: + rose_neigh_put(new_neigh); out: spin_unlock_bh(&rose_route_list_lock); spin_unlock_bh(&rose_neigh_list_lock); @@ -1190,7 +1200,7 @@ static int rose_neigh_show(struct seq_file *seq, void *v) (rose_neigh->loopback) ? "RSLOOP-0" : ax2asc(buf, &rose_neigh->callsign), rose_neigh->dev ? rose_neigh->dev->name : "???", rose_neigh->count, - rose_neigh->use, + refcount_read(&rose_neigh->use) - rose_neigh->count - 1, (rose_neigh->dce_mode) ? "DCE" : "DTE", (rose_neigh->restarted) ? "yes" : "no", ax25_display_timer(&rose_neigh->t0timer) / HZ, @@ -1295,18 +1305,22 @@ void __exit rose_rt_free(void) struct rose_neigh *s, *rose_neigh = rose_neigh_list; struct rose_node *t, *rose_node = rose_node_list; struct rose_route *u, *rose_route = rose_route_list; + int i; while (rose_neigh != NULL) { s = rose_neigh; rose_neigh = rose_neigh->next; rose_remove_neigh(s); + rose_neigh_put(s); } while (rose_node != NULL) { t = rose_node; rose_node = rose_node->next; + for (i = 0; i < t->count; i++) + rose_neigh_put(t->neighbour[i]); rose_remove_node(t); } diff --git a/net/rose/rose_timer.c b/net/rose/rose_timer.c index 020369c..bb60a16 100644 --- a/net/rose/rose_timer.c +++ b/net/rose/rose_timer.c @@ -180,7 +180,7 @@ static void rose_timer_expiry(struct timer_list *t) break; case ROSE_STATE_2: /* T3 */ - rose->neighbour->use--; + rose_neigh_put(rose->neighbour); rose_disconnect(sk, ETIMEDOUT, -1, -1); break; diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 3336dcf..568ff87 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -547,7 +547,9 @@ static void sctp_v6_from_sk(union sctp_addr *addr, struct sock *sk) { addr->v6.sin6_family = AF_INET6; addr->v6.sin6_port = 0; + addr->v6.sin6_flowinfo = 0; addr->v6.sin6_addr = sk->sk_v6_rcv_saddr; + addr->v6.sin6_scope_id = 0; } /* Initialize sk->sk_rcv_saddr from sctp_addr. */ diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index fe92e5f..dcc8a1d 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -105,12 +105,14 @@ static int virtio_transport_fill_skb(struct sk_buff *skb, size_t len, bool zcopy) { + struct msghdr *msg = info->msg; + if (zcopy) - return __zerocopy_sg_from_iter(info->msg, NULL, skb, - &info->msg->msg_iter, len, NULL); + return __zerocopy_sg_from_iter(msg, NULL, skb, + &msg->msg_iter, len, NULL); virtio_vsock_skb_put(skb, len); - return skb_copy_datagram_from_iter(skb, 0, &info->msg->msg_iter, len); + return skb_copy_datagram_from_iter_full(skb, 0, &msg->msg_iter, len); } static void virtio_transport_init_hdr(struct sk_buff *skb,