From 09e805d2570a3a94f13dd9c9ad2bcab23da76e09 Mon Sep 17 00:00:00 2001 From: Doug Berger Date: Thu, 1 Nov 2018 15:55:37 -0700 Subject: net: bcmgenet: protect stop from timeout A timing hazard exists when the network interface is stopped that allows a watchdog timeout to be processed by a separate core in parallel. This creates the potential for the timeout handler to wake the queues while the driver is shutting down, or access registers after their clocks have been removed. The more common case is that the watchdog timeout will produce a warning message which doesn't lead to a crash. The chances of this are greatly increased by the fact that bcmgenet_netif_stop stops the transmit queues which can easily precipitate a watchdog time- out because of stale trans_start data in the queues. This commit corrects the behavior by ensuring that the watchdog timeout is disabled before enterring bcmgenet_netif_stop. There are currently only two users of the bcmgenet_netif_stop function: close and suspend. The close case already handles the issue by exiting the RUNNING state before invoking the driver close service. The suspend case now performs the netif_device_detach to exit the PRESENT state before the call to bcmgenet_netif_stop rather than after it. These behaviors prevent any future scheduling of the driver timeout service during the window. The netif_tx_stop_all_queues function in bcmgenet_netif_stop is replaced with netif_tx_disable to ensure synchronization with any transmit or timeout threads that may already be executing on other cores. For symmetry, the netif_device_attach call upon resume is moved to after the call to bcmgenet_netif_start. Since it wakes the transmit queues it is not necessary to invoke netif_tx_start_all_queues from bcmgenet_netif_start so it is moved into the driver open service. Fixes: 1c1008c793fa ("net: bcmgenet: add main driver file") Signed-off-by: Doug Berger Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/genet/bcmgenet.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index 20c1681bb1af..2d6f090bf644 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -2855,7 +2855,6 @@ static void bcmgenet_netif_start(struct net_device *dev) umac_enable_set(priv, CMD_TX_EN | CMD_RX_EN, true); - netif_tx_start_all_queues(dev); bcmgenet_enable_tx_napi(priv); /* Monitor link interrupts now */ @@ -2937,6 +2936,8 @@ static int bcmgenet_open(struct net_device *dev) bcmgenet_netif_start(dev); + netif_tx_start_all_queues(dev); + return 0; err_irq1: @@ -2958,7 +2959,7 @@ static void bcmgenet_netif_stop(struct net_device *dev) struct bcmgenet_priv *priv = netdev_priv(dev); bcmgenet_disable_tx_napi(priv); - netif_tx_stop_all_queues(dev); + netif_tx_disable(dev); /* Disable MAC receive */ umac_enable_set(priv, CMD_RX_EN, false); @@ -3620,13 +3621,13 @@ static int bcmgenet_suspend(struct device *d) if (!netif_running(dev)) return 0; + netif_device_detach(dev); + bcmgenet_netif_stop(dev); if (!device_may_wakeup(d)) phy_suspend(dev->phydev); - netif_device_detach(dev); - /* Prepare the device for Wake-on-LAN and switch to the slow clock */ if (device_may_wakeup(d) && priv->wolopts) { ret = bcmgenet_power_down(priv, GENET_POWER_WOL_MAGIC); @@ -3700,8 +3701,6 @@ static int bcmgenet_resume(struct device *d) /* Always enable ring 16 - descriptor ring */ bcmgenet_enable_dma(priv, dma_ctrl); - netif_device_attach(dev); - if (!device_may_wakeup(d)) phy_resume(dev->phydev); @@ -3710,6 +3709,8 @@ static int bcmgenet_resume(struct device *d) bcmgenet_netif_start(dev); + netif_device_attach(dev); + return 0; out_clk_disable: -- cgit v1.2.3 From 7cb6a2a2c72c1ed8f42fb01f1a661281b568dead Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Thu, 1 Nov 2018 15:55:38 -0700 Subject: net: systemport: Protect stop from timeout A timing hazard exists when the network interface is stopped that allows a watchdog timeout to be processed by a separate core in parallel. This creates the potential for the timeout handler to wake the queues while the driver is shutting down, or access registers after their clocks have been removed. The more common case is that the watchdog timeout will produce a warning message which doesn't lead to a crash. The chances of this are greatly increased by the fact that bcm_sysport_netif_stop stops the transmit queues which can easily precipitate a watchdog time- out because of stale trans_start data in the queues. This commit corrects the behavior by ensuring that the watchdog timeout is disabled before enterring bcm_sysport_netif_stop. There are currently only two users of the bcm_sysport_netif_stop function: close and suspend. The close case already handles the issue by exiting the RUNNING state before invoking the driver close service. The suspend case now performs the netif_device_detach to exit the PRESENT state before the call to bcm_sysport_netif_stop rather than after it. These behaviors prevent any future scheduling of the driver timeout service during the window. The netif_tx_stop_all_queues function in bcm_sysport_netif_stop is replaced with netif_tx_disable to ensure synchronization with any transmit or timeout threads that may already be executing on other cores. For symmetry, the netif_device_attach call upon resume is moved to after the call to bcm_sysport_netif_start. Since it wakes the transmit queues it is not necessary to invoke netif_tx_start_all_queues from bcm_sysport_netif_start so it is moved into the driver open service. Fixes: 40755a0fce17 ("net: systemport: add suspend and resume support") Fixes: 80105befdb4b ("net: systemport: add Broadcom SYSTEMPORT Ethernet MAC driver") Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bcmsysport.c | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c index 4122553e224b..0e2d99c737e3 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.c +++ b/drivers/net/ethernet/broadcom/bcmsysport.c @@ -1902,9 +1902,6 @@ static void bcm_sysport_netif_start(struct net_device *dev) intrl2_1_mask_clear(priv, 0xffffffff); else intrl2_0_mask_clear(priv, INTRL2_0_TDMA_MBDONE_MASK); - - /* Last call before we start the real business */ - netif_tx_start_all_queues(dev); } static void rbuf_init(struct bcm_sysport_priv *priv) @@ -2048,6 +2045,8 @@ static int bcm_sysport_open(struct net_device *dev) bcm_sysport_netif_start(dev); + netif_tx_start_all_queues(dev); + return 0; out_clear_rx_int: @@ -2071,7 +2070,7 @@ static void bcm_sysport_netif_stop(struct net_device *dev) struct bcm_sysport_priv *priv = netdev_priv(dev); /* stop all software from updating hardware */ - netif_tx_stop_all_queues(dev); + netif_tx_disable(dev); napi_disable(&priv->napi); cancel_work_sync(&priv->dim.dim.work); phy_stop(dev->phydev); @@ -2658,12 +2657,12 @@ static int __maybe_unused bcm_sysport_suspend(struct device *d) if (!netif_running(dev)) return 0; + netif_device_detach(dev); + bcm_sysport_netif_stop(dev); phy_suspend(dev->phydev); - netif_device_detach(dev); - /* Disable UniMAC RX */ umac_enable_set(priv, CMD_RX_EN, 0); @@ -2746,8 +2745,6 @@ static int __maybe_unused bcm_sysport_resume(struct device *d) goto out_free_rx_ring; } - netif_device_attach(dev); - /* RX pipe enable */ topctrl_writel(priv, 0, RX_FLUSH_CNTL); @@ -2788,6 +2785,8 @@ static int __maybe_unused bcm_sysport_resume(struct device *d) bcm_sysport_netif_start(dev); + netif_device_attach(dev); + return 0; out_free_rx_ring: -- cgit v1.2.3 From c29734443511aa3c53844e1941805fc761aa80ba Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 31 Oct 2018 08:39:14 -0700 Subject: net/mlx4_en: use __netdev_tx_sent_queue() doorbell only depends on xmit_more and netif_tx_queue_stopped() Using __netdev_tx_sent_queue() avoids messing with BQL stop flag, and is more generic. This patch increases performance on GSO workload by keeping doorbells to the minimum required. Signed-off-by: Eric Dumazet Cc: Tariq Toukan Reviewed-by: Tariq Toukan Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/en_tx.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c index 1857ee0f0871..6f5153afcab4 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c @@ -1006,7 +1006,6 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) ring->packets++; } ring->bytes += tx_info->nr_bytes; - netdev_tx_sent_queue(ring->tx_queue, tx_info->nr_bytes); AVG_PERF_COUNTER(priv->pstats.tx_pktsz_avg, skb->len); if (tx_info->inl) @@ -1044,7 +1043,10 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) netif_tx_stop_queue(ring->tx_queue); ring->queue_stopped++; } - send_doorbell = !skb->xmit_more || netif_xmit_stopped(ring->tx_queue); + + send_doorbell = __netdev_tx_sent_queue(ring->tx_queue, + tx_info->nr_bytes, + skb->xmit_more); real_size = (real_size / 16) & 0x3f; -- cgit v1.2.3 From e8ccbb7d2f53c62e14b889faaa3f6f809b657278 Mon Sep 17 00:00:00 2001 From: Yunsheng Lin Date: Fri, 2 Nov 2018 17:47:48 +0800 Subject: net: hns3: Fix for out-of-bounds access when setting pfc back pressure The vport should be initialized to hdev->vport for each bp group, otherwise it will cause out-of-bounds access and bp setting not correct problem. [ 35.254124] BUG: KASAN: slab-out-of-bounds in hclge_pause_setup_hw+0x2a0/0x3f8 [hclge] [ 35.254126] Read of size 2 at addr ffff803b6651581a by task kworker/0:1/14 [ 35.254132] CPU: 0 PID: 14 Comm: kworker/0:1 Not tainted 4.19.0-rc7-hulk+ #85 [ 35.254133] Hardware name: Huawei D06/D06, BIOS Hisilicon D06 UEFI RC0 - B052 (V0.52) 09/14/2018 [ 35.254141] Workqueue: events work_for_cpu_fn [ 35.254144] Call trace: [ 35.254147] dump_backtrace+0x0/0x2f0 [ 35.254149] show_stack+0x24/0x30 [ 35.254154] dump_stack+0x110/0x184 [ 35.254157] print_address_description+0x168/0x2b0 [ 35.254160] kasan_report+0x184/0x310 [ 35.254162] __asan_load2+0x7c/0xa0 [ 35.254170] hclge_pause_setup_hw+0x2a0/0x3f8 [hclge] [ 35.254177] hclge_tm_init_hw+0x794/0x9f0 [hclge] [ 35.254184] hclge_tm_schd_init+0x48/0x58 [hclge] [ 35.254191] hclge_init_ae_dev+0x778/0x1168 [hclge] [ 35.254196] hnae3_register_ae_dev+0x14c/0x298 [hnae3] [ 35.254206] hns3_probe+0x88/0xa8 [hns3] [ 35.254210] local_pci_probe+0x7c/0xf0 [ 35.254212] work_for_cpu_fn+0x34/0x50 [ 35.254214] process_one_work+0x4d4/0xa38 [ 35.254216] worker_thread+0x55c/0x8d8 [ 35.254219] kthread+0x1b0/0x1b8 [ 35.254222] ret_from_fork+0x10/0x1c [ 35.254224] The buggy address belongs to the page: [ 35.254228] page:ffff7e00ed994400 count:1 mapcount:0 mapping:0000000000000000 index:0x0 compound_mapcount: 0 [ 35.273835] flags: 0xfffff8000008000(head) [ 35.282007] raw: 0fffff8000008000 dead000000000100 dead000000000200 0000000000000000 [ 35.282010] raw: 0000000000000000 0000000000000000 00000001ffffffff 0000000000000000 [ 35.282012] page dumped because: kasan: bad access detected [ 35.282014] Memory state around the buggy address: [ 35.282017] ffff803b66515700: fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe [ 35.282019] ffff803b66515780: fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe [ 35.282021] >ffff803b66515800: fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe [ 35.282022] ^ [ 35.282024] ffff803b66515880: fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe [ 35.282026] ffff803b66515900: fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe [ 35.282028] ================================================================== [ 35.282029] Disabling lock debugging due to kernel taint [ 35.282747] hclge driver initialization finished. Fixes: 67bf2541f4b9 ("net: hns3: Fixes the back pressure setting when sriov is enabled") Signed-off-by: Yunsheng Lin Signed-off-by: David S. Miller --- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c index aa5cb9834d73..494e562fe8c7 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c @@ -1168,14 +1168,14 @@ static int hclge_pfc_setup_hw(struct hclge_dev *hdev) */ static int hclge_bp_setup_hw(struct hclge_dev *hdev, u8 tc) { - struct hclge_vport *vport = hdev->vport; - u32 i, k, qs_bitmap; - int ret; + int i; for (i = 0; i < HCLGE_BP_GRP_NUM; i++) { - qs_bitmap = 0; + u32 qs_bitmap = 0; + int k, ret; for (k = 0; k < hdev->num_alloc_vport; k++) { + struct hclge_vport *vport = &hdev->vport[k]; u16 qs_id = vport->qs_offset + tc; u8 grp, sub_grp; @@ -1185,8 +1185,6 @@ static int hclge_bp_setup_hw(struct hclge_dev *hdev, u8 tc) HCLGE_BP_SUB_GRP_ID_S); if (i == grp) qs_bitmap |= (1 << sub_grp); - - vport++; } ret = hclge_tm_qs_bp_cfg(hdev, tc, i, qs_bitmap); -- cgit v1.2.3 From 92619210529a6b77b2cbedbadba3ff5eaa6e28ed Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Fri, 2 Nov 2018 16:36:24 +0100 Subject: qed: fix link config error handling gcc-8 notices that qed_mcp_get_transceiver_data() may fail to return a result to the caller: drivers/net/ethernet/qlogic/qed/qed_mcp.c: In function 'qed_mcp_trans_speed_mask': drivers/net/ethernet/qlogic/qed/qed_mcp.c:1955:2: error: 'transceiver_type' may be used uninitialized in this function [-Werror=maybe-uninitialized] When an error is returned by qed_mcp_get_transceiver_data(), we should propagate that to the caller of qed_mcp_trans_speed_mask() rather than continuing with uninitialized data. Fixes: c56a8be7e7aa ("qed: Add supported link and advertise link to display in ethtool.") Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_mcp.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c index f40f654398a0..a96364df4320 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c +++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c @@ -1944,9 +1944,12 @@ int qed_mcp_trans_speed_mask(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u32 *p_speed_mask) { u32 transceiver_type, transceiver_state; + int ret; - qed_mcp_get_transceiver_data(p_hwfn, p_ptt, &transceiver_state, - &transceiver_type); + ret = qed_mcp_get_transceiver_data(p_hwfn, p_ptt, &transceiver_state, + &transceiver_type); + if (ret) + return ret; if (qed_is_transceiver_ready(transceiver_state, transceiver_type) == false) -- cgit v1.2.3 From 96801552f846460fe9ac10f1b189602992f004e1 Mon Sep 17 00:00:00 2001 From: Shalom Toledo Date: Fri, 2 Nov 2018 19:49:15 +0000 Subject: mlxsw: spectrum: Fix IP2ME CPU policer configuration The CPU policer used to police packets being trapped via a local route (IP2ME) was incorrectly configured to police based on bytes per second instead of packets per second. Change the policer to police based on packets per second and avoid packet loss under certain circumstances. Fixes: 9148e7cf73ce ("mlxsw: spectrum: Add policers for trap groups") Signed-off-by: Shalom Toledo Signed-off-by: Ido Schimmel Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 1 - 1 file changed, 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index a2df12b79f8e..9bec940330a4 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -3568,7 +3568,6 @@ static int mlxsw_sp_cpu_policers_set(struct mlxsw_core *mlxsw_core) burst_size = 7; break; case MLXSW_REG_HTGT_TRAP_GROUP_SP_IP2ME: - is_bytes = true; rate = 4 * 1024; burst_size = 4; break; -- cgit v1.2.3 From 7131193157414ac3167d7b2f2feb4c42b415d6c5 Mon Sep 17 00:00:00 2001 From: Rasmus Villemoes Date: Mon, 5 Nov 2018 18:52:21 +0100 Subject: net: alx: make alx_drv_name static alx_drv_name is not used outside main.c, so there's no reason for it to have external linkage. Signed-off-by: Rasmus Villemoes Signed-off-by: David S. Miller --- drivers/net/ethernet/atheros/alx/alx.h | 1 - drivers/net/ethernet/atheros/alx/main.c | 2 +- 2 files changed, 1 insertion(+), 2 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/atheros/alx/alx.h b/drivers/net/ethernet/atheros/alx/alx.h index 78c5de467426..9d0e74f6b089 100644 --- a/drivers/net/ethernet/atheros/alx/alx.h +++ b/drivers/net/ethernet/atheros/alx/alx.h @@ -140,6 +140,5 @@ struct alx_priv { }; extern const struct ethtool_ops alx_ethtool_ops; -extern const char alx_drv_name[]; #endif diff --git a/drivers/net/ethernet/atheros/alx/main.c b/drivers/net/ethernet/atheros/alx/main.c index 7968c644ad86..c131cfc1b79d 100644 --- a/drivers/net/ethernet/atheros/alx/main.c +++ b/drivers/net/ethernet/atheros/alx/main.c @@ -49,7 +49,7 @@ #include "hw.h" #include "reg.h" -const char alx_drv_name[] = "alx"; +static const char alx_drv_name[] = "alx"; static void alx_free_txbuf(struct alx_tx_queue *txq, int entry) { -- cgit v1.2.3 From ce317dd9f809c8da9656c88761e30f0a82a8c2e6 Mon Sep 17 00:00:00 2001 From: Anirudh Venkataramanan Date: Fri, 26 Oct 2018 10:40:50 -0700 Subject: ice: Set carrier state and start/stop queues in rebuild Set the carrier state post rebuild by querying the link status. Also start/stop queues based on link status. Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ice/ice_main.c | 18 +++++++++++++++++- 1 file changed, 17 insertions(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 05993451147a..6d31ffb64940 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -3296,7 +3296,7 @@ static void ice_rebuild(struct ice_pf *pf) struct device *dev = &pf->pdev->dev; struct ice_hw *hw = &pf->hw; enum ice_status ret; - int err; + int err, i; if (test_bit(__ICE_DOWN, pf->state)) goto clear_recovery; @@ -3370,6 +3370,22 @@ static void ice_rebuild(struct ice_pf *pf) } ice_reset_all_vfs(pf, true); + + for (i = 0; i < pf->num_alloc_vsi; i++) { + bool link_up; + + if (!pf->vsi[i] || pf->vsi[i]->type != ICE_VSI_PF) + continue; + ice_get_link_status(pf->vsi[i]->port_info, &link_up); + if (link_up) { + netif_carrier_on(pf->vsi[i]->netdev); + netif_tx_wake_all_queues(pf->vsi[i]->netdev); + } else { + netif_carrier_off(pf->vsi[i]->netdev); + netif_tx_stop_all_queues(pf->vsi[i]->netdev); + } + } + /* if we get here, reset flow is successful */ clear_bit(__ICE_RESET_FAILED, pf->state); return; -- cgit v1.2.3 From afd9d4ab58db20029a75cf82f23b6a5641cd7d6f Mon Sep 17 00:00:00 2001 From: Anirudh Venkataramanan Date: Fri, 26 Oct 2018 10:40:51 -0700 Subject: ice: Check for reset in progress during remove The remove path does not currently check to see if a reset is in progress before proceeding. This can cause a resource collision resulting in various types of errors. Check for reset in progress and wait for a reasonable amount of time before allowing the remove to progress. Signed-off-by: Dave Ertman Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ice/ice.h | 2 ++ drivers/net/ethernet/intel/ice/ice_main.c | 6 ++++++ 2 files changed, 8 insertions(+) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index 4c4b5717a627..e5b37fa60884 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -76,6 +76,8 @@ extern const char ice_drv_ver[]; #define ICE_MIN_INTR_PER_VF (ICE_MIN_QS_PER_VF + 1) #define ICE_DFLT_INTR_PER_VF (ICE_DFLT_QS_PER_VF + 1) +#define ICE_MAX_RESET_WAIT 20 + #define ICE_VSIQF_HKEY_ARRAY_SIZE ((VSIQF_HKEY_MAX_INDEX + 1) * 4) #define ICE_DFLT_NETIF_M (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_LINK) diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 6d31ffb64940..aee22f11a41a 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -2182,6 +2182,12 @@ static void ice_remove(struct pci_dev *pdev) if (!pf) return; + for (i = 0; i < ICE_MAX_RESET_WAIT; i++) { + if (!ice_is_reset_in_progress(pf->state)) + break; + msleep(100); + } + set_bit(__ICE_DOWN, pf->state); ice_service_task_stop(pf); -- cgit v1.2.3 From 0f5d4c21a50716f8bd4e220544b82dca7408d113 Mon Sep 17 00:00:00 2001 From: Akeem G Abodunrin Date: Fri, 26 Oct 2018 10:40:52 -0700 Subject: ice: Fix dead device link issue with flow control Setting Rx or Tx pause parameter currently results in link loss on the interface, requiring the platform/host to be cold power cycled. Fix it. Signed-off-by: Akeem G Abodunrin Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ice/ice_ethtool.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c index 96923580f2a6..648acdb4c644 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c @@ -1517,10 +1517,15 @@ ice_set_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) } if (!test_bit(__ICE_DOWN, pf->state)) { - /* Give it a little more time to try to come back */ + /* Give it a little more time to try to come back. If still + * down, restart autoneg link or reinitialize the interface. + */ msleep(75); if (!test_bit(__ICE_DOWN, pf->state)) return ice_nway_reset(netdev); + + ice_down(vsi); + ice_up(vsi); } return err; -- cgit v1.2.3 From 33e055fcc26909b1d66b5d1f334aee38356068d7 Mon Sep 17 00:00:00 2001 From: Victor Raj Date: Fri, 26 Oct 2018 10:40:53 -0700 Subject: ice: Free VSI contexts during for unload In the unload path, all VSIs are freed. Also free the related VSI contexts to prevent memory leaks. Signed-off-by: Victor Raj Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ice/ice_common.c | 3 +++ drivers/net/ethernet/intel/ice/ice_switch.c | 12 ++++++++++++ drivers/net/ethernet/intel/ice/ice_switch.h | 2 ++ 3 files changed, 17 insertions(+) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c index 8cd6a2401fd9..554fd707a6d6 100644 --- a/drivers/net/ethernet/intel/ice/ice_common.c +++ b/drivers/net/ethernet/intel/ice/ice_common.c @@ -811,6 +811,9 @@ void ice_deinit_hw(struct ice_hw *hw) /* Attempt to disable FW logging before shutting down control queues */ ice_cfg_fw_log(hw, false); ice_shutdown_all_ctrlq(hw); + + /* Clear VSI contexts if not already cleared */ + ice_clear_all_vsi_ctx(hw); } /** diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c index 33403f39f1b3..40c9c6558956 100644 --- a/drivers/net/ethernet/intel/ice/ice_switch.c +++ b/drivers/net/ethernet/intel/ice/ice_switch.c @@ -347,6 +347,18 @@ static void ice_clear_vsi_ctx(struct ice_hw *hw, u16 vsi_handle) } } +/** + * ice_clear_all_vsi_ctx - clear all the VSI context entries + * @hw: pointer to the hw struct + */ +void ice_clear_all_vsi_ctx(struct ice_hw *hw) +{ + u16 i; + + for (i = 0; i < ICE_MAX_VSI; i++) + ice_clear_vsi_ctx(hw, i); +} + /** * ice_add_vsi - add VSI context to the hardware and VSI handle list * @hw: pointer to the hw struct diff --git a/drivers/net/ethernet/intel/ice/ice_switch.h b/drivers/net/ethernet/intel/ice/ice_switch.h index b88d96a1ef69..d5ef0bd58bf9 100644 --- a/drivers/net/ethernet/intel/ice/ice_switch.h +++ b/drivers/net/ethernet/intel/ice/ice_switch.h @@ -190,6 +190,8 @@ ice_update_vsi(struct ice_hw *hw, u16 vsi_handle, struct ice_vsi_ctx *vsi_ctx, struct ice_sq_cd *cd); bool ice_is_vsi_valid(struct ice_hw *hw, u16 vsi_handle); struct ice_vsi_ctx *ice_get_vsi_ctx(struct ice_hw *hw, u16 vsi_handle); +void ice_clear_all_vsi_ctx(struct ice_hw *hw); +/* Switch config */ enum ice_status ice_get_initial_sw_cfg(struct ice_hw *hw); /* Switch/bridge related commands */ -- cgit v1.2.3 From 9ecd25c26810a61b9c3abc6c73de32dca6da96e1 Mon Sep 17 00:00:00 2001 From: Anirudh Venkataramanan Date: Fri, 26 Oct 2018 10:40:54 -0700 Subject: ice: Remove duplicate addition of VLANs in replay path ice_restore_vlan and active_vlans were originally put in place to reprogram VLAN filters in the replay path. This is now done as part of the much broader VSI rebuild/replay framework. So remove both ice_restore_vlan and active_vlans Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ice/ice.h | 1 - drivers/net/ethernet/intel/ice/ice_main.c | 42 ++++-------------------- drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c | 2 -- 3 files changed, 6 insertions(+), 39 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index e5b37fa60884..1639e955f158 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -191,7 +191,6 @@ struct ice_vsi { u64 tx_linearize; DECLARE_BITMAP(state, __ICE_STATE_NBITS); DECLARE_BITMAP(flags, ICE_VSI_FLAG_NBITS); - unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)]; unsigned int current_netdev_flags; u32 tx_restart; u32 tx_busy; diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index aee22f11a41a..338abb1b9233 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -1622,7 +1622,6 @@ static int ice_vlan_rx_add_vid(struct net_device *netdev, { struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; - int ret; if (vid >= VLAN_N_VID) { netdev_err(netdev, "VLAN id requested %d is out of range %d\n", @@ -1635,7 +1634,8 @@ static int ice_vlan_rx_add_vid(struct net_device *netdev, /* Enable VLAN pruning when VLAN 0 is added */ if (unlikely(!vid)) { - ret = ice_cfg_vlan_pruning(vsi, true); + int ret = ice_cfg_vlan_pruning(vsi, true); + if (ret) return ret; } @@ -1644,12 +1644,7 @@ static int ice_vlan_rx_add_vid(struct net_device *netdev, * needed to continue allowing all untagged packets since VLAN prune * list is applied to all packets by the switch */ - ret = ice_vsi_add_vlan(vsi, vid); - - if (!ret) - set_bit(vid, vsi->active_vlans); - - return ret; + return ice_vsi_add_vlan(vsi, vid); } /** @@ -1677,8 +1672,6 @@ static int ice_vlan_rx_kill_vid(struct net_device *netdev, if (status) return status; - clear_bit(vid, vsi->active_vlans); - /* Disable VLAN pruning when VLAN 0 is removed */ if (unlikely(!vid)) status = ice_cfg_vlan_pruning(vsi, false); @@ -2515,31 +2508,6 @@ static int ice_vsi_vlan_setup(struct ice_vsi *vsi) return ret; } -/** - * ice_restore_vlan - Reinstate VLANs when vsi/netdev comes back up - * @vsi: the VSI being brought back up - */ -static int ice_restore_vlan(struct ice_vsi *vsi) -{ - int err; - u16 vid; - - if (!vsi->netdev) - return -EINVAL; - - err = ice_vsi_vlan_setup(vsi); - if (err) - return err; - - for_each_set_bit(vid, vsi->active_vlans, VLAN_N_VID) { - err = ice_vlan_rx_add_vid(vsi->netdev, htons(ETH_P_8021Q), vid); - if (err) - break; - } - - return err; -} - /** * ice_vsi_cfg - Setup the VSI * @vsi: the VSI being configured @@ -2552,7 +2520,9 @@ static int ice_vsi_cfg(struct ice_vsi *vsi) if (vsi->netdev) { ice_set_rx_mode(vsi->netdev); - err = ice_restore_vlan(vsi); + + err = ice_vsi_vlan_setup(vsi); + if (err) return err; } diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c index 45f10f8f01dc..9576b958622b 100644 --- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c @@ -2171,7 +2171,6 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v) if (!ice_vsi_add_vlan(vsi, vid)) { vf->num_vlan++; - set_bit(vid, vsi->active_vlans); /* Enable VLAN pruning when VLAN 0 is added */ if (unlikely(!vid)) @@ -2190,7 +2189,6 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v) */ if (!ice_vsi_kill_vlan(vsi, vid)) { vf->num_vlan--; - clear_bit(vid, vsi->active_vlans); /* Disable VLAN pruning when removing VLAN 0 */ if (unlikely(!vid)) -- cgit v1.2.3 From 58297dd133f64ea028e7d52dd00cd9ff8aa4479f Mon Sep 17 00:00:00 2001 From: Md Fahad Iqbal Polash Date: Fri, 26 Oct 2018 10:40:55 -0700 Subject: ice: Fix flags for port VLAN According to the spec, whenever insert PVID field is set, the VLAN driver insertion mode should be set to 01b which isn't done currently. Fix it. Signed-off-by: Md Fahad Iqbal Polash Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c index 9576b958622b..e71065f9d391 100644 --- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c @@ -348,7 +348,7 @@ static int ice_vsi_set_pvid(struct ice_vsi *vsi, u16 vid) struct ice_vsi_ctx ctxt = { 0 }; enum ice_status status; - ctxt.info.vlan_flags = ICE_AQ_VSI_VLAN_MODE_TAGGED | + ctxt.info.vlan_flags = ICE_AQ_VSI_VLAN_MODE_UNTAGGED | ICE_AQ_VSI_PVLAN_INSERT_PVID | ICE_AQ_VSI_VLAN_EMOD_STR; ctxt.info.pvid = cpu_to_le16(vid); -- cgit v1.2.3 From 31082519c11b01fe1fb6dd512055f252812c1508 Mon Sep 17 00:00:00 2001 From: Anirudh Venkataramanan Date: Fri, 26 Oct 2018 10:40:56 -0700 Subject: ice: Fix typo in error message Print should say "Enabling" instead of "Enaabling" Signed-off-by: Akeem G Abodunrin Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ice/ice_lib.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c index 5bacad01f0c9..c604a44c8cfb 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_lib.c @@ -1997,7 +1997,7 @@ int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena) status = ice_update_vsi(&vsi->back->hw, vsi->idx, ctxt, NULL); if (status) { netdev_err(vsi->netdev, "%sabling VLAN pruning on VSI handle: %d, VSI HW ID: %d failed, err = %d, aq_err = %d\n", - ena ? "Ena" : "Dis", vsi->idx, vsi->vsi_num, status, + ena ? "En" : "Dis", vsi->idx, vsi->vsi_num, status, vsi->back->hw.adminq.sq_last_status); goto err_out; } -- cgit v1.2.3 From 25525b69bb44a628841492f44a5a8e74f34724f4 Mon Sep 17 00:00:00 2001 From: Dave Ertman Date: Fri, 26 Oct 2018 10:40:57 -0700 Subject: ice: Fix napi delete calls for remove In the remove path, the vsi->netdev is being set to NULL before the call to free vectors. This is causing the netif_napi_del call to never be made. Add a call to ice_napi_del to the same location as the calls to unregister_netdev and just prior to them. This will use the reverse flow as the register and netif_napi_add calls. Signed-off-by: Dave Ertman Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ice/ice.h | 1 + drivers/net/ethernet/intel/ice/ice_lib.c | 1 + drivers/net/ethernet/intel/ice/ice_main.c | 2 +- 3 files changed, 3 insertions(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index 1639e955f158..b8548370f1c7 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -370,5 +370,6 @@ int ice_set_rss(struct ice_vsi *vsi, u8 *seed, u8 *lut, u16 lut_size); int ice_get_rss(struct ice_vsi *vsi, u8 *seed, u8 *lut, u16 lut_size); void ice_fill_rss_lut(u8 *lut, u16 rss_table_size, u16 rss_size); void ice_print_link_msg(struct ice_vsi *vsi, bool isup); +void ice_napi_del(struct ice_vsi *vsi); #endif /* _ICE_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c index c604a44c8cfb..1041fa2a7767 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_lib.c @@ -2458,6 +2458,7 @@ int ice_vsi_release(struct ice_vsi *vsi) * on this wq */ if (vsi->netdev && !ice_is_reset_in_progress(pf->state)) { + ice_napi_del(vsi); unregister_netdev(vsi->netdev); free_netdev(vsi->netdev); vsi->netdev = NULL; diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 338abb1b9233..82f49dbd762c 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -1465,7 +1465,7 @@ skip_req_irq: * ice_napi_del - Remove NAPI handler for the VSI * @vsi: VSI for which NAPI handler is to be removed */ -static void ice_napi_del(struct ice_vsi *vsi) +void ice_napi_del(struct ice_vsi *vsi) { int v_idx; -- cgit v1.2.3 From c585ea42ec75e8d3afa278b7095d9f0dd6ee515b Mon Sep 17 00:00:00 2001 From: Brett Creeley Date: Fri, 26 Oct 2018 10:40:58 -0700 Subject: ice: Fix tx_timeout in PF driver Prior to this commit the driver was running into tx_timeouts when a queue was stressed enough. This was happening because the HW tail and SW tail (NTU) were incorrectly out of sync. Consequently this was causing the HW head to collide with the HW tail, which to the hardware means that all descriptors posted for Tx have been processed. Due to the Tx logic used in the driver SW tail and HW tail are allowed to be out of sync. This is done as an optimization because it allows the driver to write HW tail as infrequently as possible, while still updating the SW tail index to keep track. However, there are situations where this results in the tail never getting updated, resulting in Tx timeouts. Tx HW tail write condition: if (netif_xmit_stopped(txring_txq(tx_ring) || !skb->xmit_more) writel(sw_tail, tx_ring->tail); An issue was found in the Tx logic that was causing the afore mentioned condition for updating HW tail to never happen, causing tx_timeouts. In ice_xmit_frame_ring we calculate how many descriptors we need for the Tx transaction based on the skb the kernel hands us. This is then passed into ice_maybe_stop_tx along with some extra padding to determine if we have enough descriptors available for this transaction. If we don't then we return -EBUSY to the stack, otherwise we move on and eventually prepare the Tx descriptors accordingly in ice_tx_map and set next_to_watch. In ice_tx_map we make another call to ice_maybe_stop_tx with a value of MAX_SKB_FRAGS + 4. The key here is that this value is possibly less than the value we sent in the first call to ice_maybe_stop_tx in ice_xmit_frame_ring. Now, if the number of unused descriptors is between MAX_SKB_FRAGS + 4 and the value used in the first call to ice_maybe_stop_tx in ice_xmit_frame_ring then we do not update the HW tail because of the "Tx HW tail write condition" above. This is because in ice_maybe_stop_tx we return success from ice_maybe_stop_tx instead of calling __ice_maybe_stop_tx and subsequently calling netif_stop_subqueue, which sets the __QUEUE_STATE_DEV_XOFF bit. This bit is then checked in the "Tx HW tail write condition" by calling netif_xmit_stopped and subsequently updating HW tail if the afore mentioned bit is set. In ice_clean_tx_irq, if next_to_watch is not NULL, we end up cleaning the descriptors that HW sets the DD bit on and we have the budget. The HW head will eventually run into the HW tail in response to the description in the paragraph above. The next time through ice_xmit_frame_ring we make the initial call to ice_maybe_stop_tx with another skb from the stack. This time we do not have enough descriptors available and we return NETDEV_TX_BUSY to the stack and end up setting next_to_watch to NULL. This is where we are stuck. In ice_clean_tx_irq we never clean anything because next_to_watch is always NULL and in ice_xmit_frame_ring we never update HW tail because we already return NETDEV_TX_BUSY to the stack and eventually we hit a tx_timeout. This issue was fixed by making sure that the second call to ice_maybe_stop_tx in ice_tx_map is passed a value that is >= the value that was used on the initial call to ice_maybe_stop_tx in ice_xmit_frame_ring. This was done by adding the following defines to make the logic more clear and to reduce the chance of mucking this up again: ICE_CACHE_LINE_BYTES 64 ICE_DESCS_PER_CACHE_LINE (ICE_CACHE_LINE_BYTES / \ sizeof(struct ice_tx_desc)) ICE_DESCS_FOR_CTX_DESC 1 ICE_DESCS_FOR_SKB_DATA_PTR 1 The ICE_CACHE_LINE_BYTES being 64 is an assumption being made so we don't have to figure this out on every pass through the Tx path. Instead I added a sanity check in ice_probe to verify cache line size and print a message if it's not 64 Bytes. This will make it easier to file issues if they are seen when the cache line size is not 64 Bytes when reading from the GLPCI_CNF2 register. Signed-off-by: Brett Creeley Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ice/ice_hw_autogen.h | 2 ++ drivers/net/ethernet/intel/ice/ice_main.c | 18 ++++++++++++++++++ drivers/net/ethernet/intel/ice/ice_txrx.c | 9 +++++---- drivers/net/ethernet/intel/ice/ice_txrx.h | 17 +++++++++++++++-- 4 files changed, 40 insertions(+), 6 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h index 5fdea6ec7675..596b9fb1c510 100644 --- a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h +++ b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h @@ -242,6 +242,8 @@ #define GLNVM_ULD 0x000B6008 #define GLNVM_ULD_CORER_DONE_M BIT(3) #define GLNVM_ULD_GLOBR_DONE_M BIT(4) +#define GLPCI_CNF2 0x000BE004 +#define GLPCI_CNF2_CACHELINE_SIZE_M BIT(1) #define PF_FUNC_RID 0x0009E880 #define PF_FUNC_RID_FUNC_NUM_S 0 #define PF_FUNC_RID_FUNC_NUM_M ICE_M(0x7, 0) diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 82f49dbd762c..333312a1d595 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -1994,6 +1994,22 @@ static int ice_init_interrupt_scheme(struct ice_pf *pf) return 0; } +/** + * ice_verify_cacheline_size - verify driver's assumption of 64 Byte cache lines + * @pf: pointer to the PF structure + * + * There is no error returned here because the driver should be able to handle + * 128 Byte cache lines, so we only print a warning in case issues are seen, + * specifically with Tx. + */ +static void ice_verify_cacheline_size(struct ice_pf *pf) +{ + if (rd32(&pf->hw, GLPCI_CNF2) & GLPCI_CNF2_CACHELINE_SIZE_M) + dev_warn(&pf->pdev->dev, + "%d Byte cache line assumption is invalid, driver may have Tx timeouts!\n", + ICE_CACHE_LINE_BYTES); +} + /** * ice_probe - Device initialization routine * @pdev: PCI device information struct @@ -2144,6 +2160,8 @@ static int ice_probe(struct pci_dev *pdev, /* since everything is good, start the service timer */ mod_timer(&pf->serv_tmr, round_jiffies(jiffies + pf->serv_tmr_period)); + ice_verify_cacheline_size(pf); + return 0; err_alloc_sw_unroll: diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index 5dae968d853e..3387c67c848d 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -1556,15 +1556,15 @@ int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off) * magnitude greater than our largest possible GSO size. * * This would then be implemented as: - * return (((size >> 12) * 85) >> 8) + 1; + * return (((size >> 12) * 85) >> 8) + ICE_DESCS_FOR_SKB_DATA_PTR; * * Since multiplication and division are commutative, we can reorder * operations into: - * return ((size * 85) >> 20) + 1; + * return ((size * 85) >> 20) + ICE_DESCS_FOR_SKB_DATA_PTR; */ static unsigned int ice_txd_use_count(unsigned int size) { - return ((size * 85) >> 20) + 1; + return ((size * 85) >> 20) + ICE_DESCS_FOR_SKB_DATA_PTR; } /** @@ -1706,7 +1706,8 @@ ice_xmit_frame_ring(struct sk_buff *skb, struct ice_ring *tx_ring) * + 1 desc for context descriptor, * otherwise try next time */ - if (ice_maybe_stop_tx(tx_ring, count + 4 + 1)) { + if (ice_maybe_stop_tx(tx_ring, count + ICE_DESCS_PER_CACHE_LINE + + ICE_DESCS_FOR_CTX_DESC)) { tx_ring->tx_stats.tx_busy++; return NETDEV_TX_BUSY; } diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h index 1d0f58bd389b..75d0eaf6c9dd 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.h +++ b/drivers/net/ethernet/intel/ice/ice_txrx.h @@ -22,8 +22,21 @@ #define ICE_RX_BUF_WRITE 16 /* Must be power of 2 */ #define ICE_MAX_TXQ_PER_TXQG 128 -/* Tx Descriptors needed, worst case */ -#define DESC_NEEDED (MAX_SKB_FRAGS + 4) +/* We are assuming that the cache line is always 64 Bytes here for ice. + * In order to make sure that is a correct assumption there is a check in probe + * to print a warning if the read from GLPCI_CNF2 tells us that the cache line + * size is 128 bytes. We do it this way because we do not want to read the + * GLPCI_CNF2 register or a variable containing the value on every pass through + * the Tx path. + */ +#define ICE_CACHE_LINE_BYTES 64 +#define ICE_DESCS_PER_CACHE_LINE (ICE_CACHE_LINE_BYTES / \ + sizeof(struct ice_tx_desc)) +#define ICE_DESCS_FOR_CTX_DESC 1 +#define ICE_DESCS_FOR_SKB_DATA_PTR 1 +/* Tx descriptors needed, worst case */ +#define DESC_NEEDED (MAX_SKB_FRAGS + ICE_DESCS_FOR_CTX_DESC + \ + ICE_DESCS_PER_CACHE_LINE + ICE_DESCS_FOR_SKB_DATA_PTR) #define ICE_DESC_UNUSED(R) \ ((((R)->next_to_clean > (R)->next_to_use) ? 0 : (R)->count) + \ (R)->next_to_clean - (R)->next_to_use - 1) -- cgit v1.2.3 From d944b46992f8e99b6bdc721e44b02e5ca294fa2b Mon Sep 17 00:00:00 2001 From: Brett Creeley Date: Fri, 26 Oct 2018 10:40:59 -0700 Subject: ice: Fix the bytecount sent to netdev_tx_sent_queue Currently if the driver does a TSO offload the bytecount sent to netdev_tx_sent_queue will be incorrect. This is because in ice_tso we overwrite the initial value that we set in ice_tx_map. This creates a mismatch between the Tx and Tx clean flow. In the Tx clean flow we calculate the bytecount (called total_bytes) as we clean the descriptors so the value used in the Tx clean path is correct. Fix this by using += in ice_tso instead of =. This fixes the mismatch in bytecount mentioned above. Signed-off-by: Brett Creeley Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ice/ice_txrx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index 3387c67c848d..fe5bbabbb41e 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -1520,7 +1520,7 @@ int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off) /* update gso_segs and bytecount */ first->gso_segs = skb_shinfo(skb)->gso_segs; - first->bytecount = (first->gso_segs - 1) * off->header_len; + first->bytecount += (first->gso_segs - 1) * off->header_len; cd_tso_len = skb->len - off->header_len; cd_mss = skb_shinfo(skb)->gso_size; -- cgit v1.2.3 From 4c9b658eeaefedd402a59e858d8ac3bfdf6153e3 Mon Sep 17 00:00:00 2001 From: Miroslav Lichvar Date: Fri, 26 Oct 2018 19:13:00 +0200 Subject: igb: shorten maximum PHC timecounter update interval The timecounter needs to be updated at least once per ~550 seconds in order to avoid a 40-bit SYSTIM timestamp to be misinterpreted as an old timestamp. Since commit 500462a9de65 ("timers: Switch to a non-cascading wheel"), scheduling of delayed work seems to be less accurate and a requested delay of 540 seconds may actually be longer than 550 seconds. Also, the PHC may be adjusted to run up to 6% faster than real time and the system clock up to 10% slower. Shorten the delay to 360 seconds to be sure the timecounter is updated in time. This fixes an issue with HW timestamps on 82580/I350/I354 being off by ~1100 seconds for few seconds every ~9 minutes. Cc: Thomas Gleixner Signed-off-by: Miroslav Lichvar Acked-by: Jacob Keller Acked-by: Richard Cochran Tested-by: Aaron Brown Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/igb/igb_ptp.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/intel/igb/igb_ptp.c b/drivers/net/ethernet/intel/igb/igb_ptp.c index 29ced6b74d36..2b95dc9c7a6a 100644 --- a/drivers/net/ethernet/intel/igb/igb_ptp.c +++ b/drivers/net/ethernet/intel/igb/igb_ptp.c @@ -53,13 +53,15 @@ * 2^40 * 10^-9 / 60 = 18.3 minutes. * * SYSTIM is converted to real time using a timecounter. As - * timecounter_cyc2time() allows old timestamps, the timecounter - * needs to be updated at least once per half of the SYSTIM interval. - * Scheduling of delayed work is not very accurate, so we aim for 8 - * minutes to be sure the actual interval is shorter than 9.16 minutes. + * timecounter_cyc2time() allows old timestamps, the timecounter needs + * to be updated at least once per half of the SYSTIM interval. + * Scheduling of delayed work is not very accurate, and also the NIC + * clock can be adjusted to run up to 6% faster and the system clock + * up to 10% slower, so we aim for 6 minutes to be sure the actual + * interval in the NIC time is shorter than 9.16 minutes. */ -#define IGB_SYSTIM_OVERFLOW_PERIOD (HZ * 60 * 8) +#define IGB_SYSTIM_OVERFLOW_PERIOD (HZ * 60 * 6) #define IGB_PTP_TX_TIMEOUT (HZ * 15) #define INCPERIOD_82576 BIT(E1000_TIMINCA_16NS_SHIFT) #define INCVALUE_82576_MASK GENMASK(E1000_TIMINCA_16NS_SHIFT - 1, 0) -- cgit v1.2.3 From ffe498237b36ee42624e139b21efa05da4ff1f48 Mon Sep 17 00:00:00 2001 From: Chinh T Cao Date: Mon, 5 Nov 2018 12:18:35 -0800 Subject: ice: Change req_speeds to be u16 Since the req_speeds field in struct ice_link_status is a u8, req_speeds & ICE_AQ_LINK_SPEED_40GB always returns 0. This was caught by a coverity scan. Fix this by changing req_speeds to be u16. Reported-by: Bruce Allan Signed-off-by: Chinh T Cao Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ice/ice_type.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h index 12f9432abf11..f4dbc81c1988 100644 --- a/drivers/net/ethernet/intel/ice/ice_type.h +++ b/drivers/net/ethernet/intel/ice/ice_type.h @@ -92,12 +92,12 @@ struct ice_link_status { u64 phy_type_low; u16 max_frame_size; u16 link_speed; + u16 req_speeds; u8 lse_ena; /* Link Status Event notification */ u8 link_info; u8 an_info; u8 ext_info; u8 pacing; - u8 req_speeds; /* Refer to #define from module_type[ICE_MODULE_TYPE_TOTAL_BYTE] of * ice_aqc_get_phy_caps structure */ -- cgit v1.2.3 From ba766b8b99c30ad3c55ed8cf224d1185ecff1476 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Mon, 29 Oct 2018 10:52:42 -0700 Subject: i40e: restore NETIF_F_GSO_IPXIP[46] to netdev features Since commit bacd75cfac8a ("i40e/i40evf: Add capability exchange for outer checksum", 2017-04-06) the i40e driver has not reported support for IP-in-IP offloads. This likely occurred due to a bad rebase, as the commit extracts hw_enc_features into its own variable. As part of this change, it dropped the NETIF_F_FSO_IPXIP flags from the netdev->hw_enc_features. This was unfortunately not caught during code review. Fix this by adding back the missing feature flags. For reference, NETIF_F_GSO_IPXIP4 was added in commit 7e13318daa4a ("net: define gso types for IPx over IPv4 and IPv6", 2016-05-20), replacing NETIF_F_GSO_IPIP and NETIF_F_GSO_SIT. NETIF_F_GSO_IPXIP6 was added in commit bf2d1df39502 ("intel: Add support for IPv6 IP-in-IP offload", 2016-05-20). Signed-off-by: Jacob Keller Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/i40e/i40e_main.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index bc71a21c1dc2..3ff5ee49818b 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -12249,6 +12249,8 @@ static int i40e_config_netdev(struct i40e_vsi *vsi) NETIF_F_GSO_GRE | NETIF_F_GSO_GRE_CSUM | NETIF_F_GSO_PARTIAL | + NETIF_F_GSO_IPXIP4 | + NETIF_F_GSO_IPXIP6 | NETIF_F_GSO_UDP_TUNNEL | NETIF_F_GSO_UDP_TUNNEL_CSUM | NETIF_F_SCTP_CRC | -- cgit v1.2.3 From d5596fd46770550873ab4c02bcb69f83d3d63f63 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Mon, 29 Oct 2018 10:52:43 -0700 Subject: i40e: enable NETIF_F_NTUPLE and NETIF_F_HW_TC at driver load The assignment of the feature flag NETIF_F_NTUPLE and NETIF_F_HW_TC occurs prior to the initial setup of the local hw_features variable. This means the features are set as user-changeable, but are not set in the currently active feature list. This results in the features being disabled at the driver's initial load. Move the assignment after the initial assignment of hw_features, and assign to the local variable. This ensures that NETIF_F_NTUPLE and NETIF_F_HW_TC are marked as user-changeable, and also enables them by default when the driver loads. Signed-off-by: Jacob Keller Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/i40e/i40e_main.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 3ff5ee49818b..21c2688d6308 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -12268,13 +12268,13 @@ static int i40e_config_netdev(struct i40e_vsi *vsi) /* record features VLANs can make use of */ netdev->vlan_features |= hw_enc_features | NETIF_F_TSO_MANGLEID; - if (!(pf->flags & I40E_FLAG_MFP_ENABLED)) - netdev->hw_features |= NETIF_F_NTUPLE | NETIF_F_HW_TC; - hw_features = hw_enc_features | NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX; + if (!(pf->flags & I40E_FLAG_MFP_ENABLED)) + hw_features |= NETIF_F_NTUPLE | NETIF_F_HW_TC; + netdev->hw_features |= hw_features; netdev->features |= hw_features | NETIF_F_HW_VLAN_CTAG_FILTER; -- cgit v1.2.3 From e84b47941e15e6666afb8ee8b21d1c3fc1a013af Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Micha=C5=82=20Miros=C5=82aw?= Date: Wed, 7 Nov 2018 17:50:52 +0100 Subject: ibmvnic: fix accelerated VLAN handling MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Don't request tag insertion when it isn't present in outgoing skb. Signed-off-by: Michał Mirosław Signed-off-by: David S. Miller --- drivers/net/ethernet/ibm/ibmvnic.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index 7893beffcc71..c9d5d0a7fbf1 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -1545,7 +1545,7 @@ static netdev_tx_t ibmvnic_xmit(struct sk_buff *skb, struct net_device *netdev) tx_crq.v1.sge_len = cpu_to_be32(skb->len); tx_crq.v1.ioba = cpu_to_be64(data_dma_addr); - if (adapter->vlan_header_insertion) { + if (adapter->vlan_header_insertion && skb_vlan_tag_present(skb)) { tx_crq.v1.flags2 |= IBMVNIC_TX_VLAN_INSERT; tx_crq.v1.vlan_id = cpu_to_be16(skb->vlan_tci); } -- cgit v1.2.3 From b25ddb00bc1b96613edcb525f19203a7d1405fce Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Micha=C5=82=20Miros=C5=82aw?= Date: Wed, 7 Nov 2018 17:50:53 +0100 Subject: qlcnic: remove assumption that vlan_tci != 0 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit VLAN.TCI == 0 is perfectly valid (802.1p), so allow it to be accelerated. Signed-off-by: Michał Mirosław Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c index 9647578cbe6a..14f26bf3b388 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c @@ -459,7 +459,7 @@ static int qlcnic_tx_pkt(struct qlcnic_adapter *adapter, struct cmd_desc_type0 *first_desc, struct sk_buff *skb, struct qlcnic_host_tx_ring *tx_ring) { - u8 l4proto, opcode = 0, hdr_len = 0; + u8 l4proto, opcode = 0, hdr_len = 0, tag_vlan = 0; u16 flags = 0, vlan_tci = 0; int copied, offset, copy_len, size; struct cmd_desc_type0 *hwdesc; @@ -472,14 +472,16 @@ static int qlcnic_tx_pkt(struct qlcnic_adapter *adapter, flags = QLCNIC_FLAGS_VLAN_TAGGED; vlan_tci = ntohs(vh->h_vlan_TCI); protocol = ntohs(vh->h_vlan_encapsulated_proto); + tag_vlan = 1; } else if (skb_vlan_tag_present(skb)) { flags = QLCNIC_FLAGS_VLAN_OOB; vlan_tci = skb_vlan_tag_get(skb); + tag_vlan = 1; } if (unlikely(adapter->tx_pvid)) { - if (vlan_tci && !(adapter->flags & QLCNIC_TAGGING_ENABLED)) + if (tag_vlan && !(adapter->flags & QLCNIC_TAGGING_ENABLED)) return -EIO; - if (vlan_tci && (adapter->flags & QLCNIC_TAGGING_ENABLED)) + if (tag_vlan && (adapter->flags & QLCNIC_TAGGING_ENABLED)) goto set_flags; flags = QLCNIC_FLAGS_VLAN_OOB; -- cgit v1.2.3 From e12c225258f2584906765234ca6db4ad4c618192 Mon Sep 17 00:00:00 2001 From: Huazhong Tan Date: Thu, 8 Nov 2018 10:13:19 +0800 Subject: net: hns3: bugfix for not checking return value hns3_reset_notify_init_enet() only return error early if the return value of hns3_restore_vlan() is not 0. This patch adds checking for the return value of hns3_restore_vlan. Fixes: 7fa6be4fd2f6 ("net: hns3: fix incorrect return value/type of some functions") Signed-off-by: Huazhong Tan Signed-off-by: David S. Miller --- drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index 3f96aa30068e..20fcf0d1c2ce 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -3760,7 +3760,8 @@ static int hns3_reset_notify_init_enet(struct hnae3_handle *handle) /* Hardware table is only clear when pf resets */ if (!(handle->flags & HNAE3_SUPPORT_VF)) { ret = hns3_restore_vlan(netdev); - return ret; + if (ret) + return ret; } ret = hns3_restore_fd_rules(netdev); -- cgit v1.2.3 From 39477551df940ddb1339203817de04f5caaacf7a Mon Sep 17 00:00:00 2001 From: Denis Bolotin Date: Thu, 8 Nov 2018 16:46:08 +0200 Subject: qed: Fix memory/entry leak in qed_init_sp_request() Free the allocated SPQ entry or return the acquired SPQ entry to the free list in error flows. Signed-off-by: Denis Bolotin Signed-off-by: Michal Kalderon Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_sp_commands.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c index 77b6248ad3b9..e86a1ea23613 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c @@ -80,7 +80,7 @@ int qed_sp_init_request(struct qed_hwfn *p_hwfn, case QED_SPQ_MODE_BLOCK: if (!p_data->p_comp_data) - return -EINVAL; + goto err; p_ent->comp_cb.cookie = p_data->p_comp_data->cookie; break; @@ -95,7 +95,7 @@ int qed_sp_init_request(struct qed_hwfn *p_hwfn, default: DP_NOTICE(p_hwfn, "Unknown SPQE completion mode %d\n", p_ent->comp_mode); - return -EINVAL; + goto err; } DP_VERBOSE(p_hwfn, QED_MSG_SPQ, @@ -109,6 +109,18 @@ int qed_sp_init_request(struct qed_hwfn *p_hwfn, memset(&p_ent->ramrod, 0, sizeof(p_ent->ramrod)); return 0; + +err: + /* qed_spq_get_entry() can either get an entry from the free_pool, + * or, if no entries are left, allocate a new entry and add it to + * the unlimited_pending list. + */ + if (p_ent->queue == &p_hwfn->p_spq->unlimited_pending) + kfree(p_ent); + else + qed_spq_return_entry(p_hwfn, p_ent); + + return -EINVAL; } static enum tunnel_clss qed_tunn_clss_to_fw_clss(u8 type) -- cgit v1.2.3 From 2632f22ebd08da249c2017962a199a0cfb2324bf Mon Sep 17 00:00:00 2001 From: Denis Bolotin Date: Thu, 8 Nov 2018 16:46:09 +0200 Subject: qed: Fix blocking/unlimited SPQ entries leak When there are no SPQ entries left in the free_pool, new entries are allocated and are added to the unlimited list. When an entry in the pool is available, the content is copied from the original entry, and the new entry is sent to the device. qed_spq_post() is not aware of that, so the additional entry is stored in the original entry as p_post_ent, which can later be returned to the pool. Signed-off-by: Denis Bolotin Signed-off-by: Michal Kalderon Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_sp.h | 3 ++ drivers/net/ethernet/qlogic/qed/qed_spq.c | 57 ++++++++++++++++--------------- 2 files changed, 33 insertions(+), 27 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp.h b/drivers/net/ethernet/qlogic/qed/qed_sp.h index e95431f6acd4..04259df8a5c2 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sp.h +++ b/drivers/net/ethernet/qlogic/qed/qed_sp.h @@ -167,6 +167,9 @@ struct qed_spq_entry { enum spq_mode comp_mode; struct qed_spq_comp_cb comp_cb; struct qed_spq_comp_done comp_done; /* SPQ_MODE_EBLOCK */ + + /* Posted entry for unlimited list entry in EBLOCK mode */ + struct qed_spq_entry *post_ent; }; struct qed_eq { diff --git a/drivers/net/ethernet/qlogic/qed/qed_spq.c b/drivers/net/ethernet/qlogic/qed/qed_spq.c index c4a6274dd625..c1a81ec0524b 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_spq.c +++ b/drivers/net/ethernet/qlogic/qed/qed_spq.c @@ -685,6 +685,8 @@ static int qed_spq_add_entry(struct qed_hwfn *p_hwfn, /* EBLOCK responsible to free the allocated p_ent */ if (p_ent->comp_mode != QED_SPQ_MODE_EBLOCK) kfree(p_ent); + else + p_ent->post_ent = p_en2; p_ent = p_en2; } @@ -767,6 +769,25 @@ static int qed_spq_pend_post(struct qed_hwfn *p_hwfn) SPQ_HIGH_PRI_RESERVE_DEFAULT); } +/* Avoid overriding of SPQ entries when getting out-of-order completions, by + * marking the completions in a bitmap and increasing the chain consumer only + * for the first successive completed entries. + */ +static void qed_spq_comp_bmap_update(struct qed_hwfn *p_hwfn, __le16 echo) +{ + u16 pos = le16_to_cpu(echo) % SPQ_RING_SIZE; + struct qed_spq *p_spq = p_hwfn->p_spq; + + __set_bit(pos, p_spq->p_comp_bitmap); + while (test_bit(p_spq->comp_bitmap_idx, + p_spq->p_comp_bitmap)) { + __clear_bit(p_spq->comp_bitmap_idx, + p_spq->p_comp_bitmap); + p_spq->comp_bitmap_idx++; + qed_chain_return_produced(&p_spq->chain); + } +} + int qed_spq_post(struct qed_hwfn *p_hwfn, struct qed_spq_entry *p_ent, u8 *fw_return_code) { @@ -824,11 +845,12 @@ int qed_spq_post(struct qed_hwfn *p_hwfn, p_ent->queue == &p_spq->unlimited_pending); if (p_ent->queue == &p_spq->unlimited_pending) { - /* This is an allocated p_ent which does not need to - * return to pool. - */ + struct qed_spq_entry *p_post_ent = p_ent->post_ent; + kfree(p_ent); - return rc; + + /* Return the entry which was actually posted */ + p_ent = p_post_ent; } if (rc) @@ -842,7 +864,7 @@ int qed_spq_post(struct qed_hwfn *p_hwfn, spq_post_fail2: spin_lock_bh(&p_spq->lock); list_del(&p_ent->list); - qed_chain_return_produced(&p_spq->chain); + qed_spq_comp_bmap_update(p_hwfn, p_ent->elem.hdr.echo); spq_post_fail: /* return to the free pool */ @@ -874,25 +896,8 @@ int qed_spq_completion(struct qed_hwfn *p_hwfn, spin_lock_bh(&p_spq->lock); list_for_each_entry_safe(p_ent, tmp, &p_spq->completion_pending, list) { if (p_ent->elem.hdr.echo == echo) { - u16 pos = le16_to_cpu(echo) % SPQ_RING_SIZE; - list_del(&p_ent->list); - - /* Avoid overriding of SPQ entries when getting - * out-of-order completions, by marking the completions - * in a bitmap and increasing the chain consumer only - * for the first successive completed entries. - */ - __set_bit(pos, p_spq->p_comp_bitmap); - - while (test_bit(p_spq->comp_bitmap_idx, - p_spq->p_comp_bitmap)) { - __clear_bit(p_spq->comp_bitmap_idx, - p_spq->p_comp_bitmap); - p_spq->comp_bitmap_idx++; - qed_chain_return_produced(&p_spq->chain); - } - + qed_spq_comp_bmap_update(p_hwfn, echo); p_spq->comp_count++; found = p_ent; break; @@ -931,11 +936,9 @@ int qed_spq_completion(struct qed_hwfn *p_hwfn, QED_MSG_SPQ, "Got a completion without a callback function\n"); - if ((found->comp_mode != QED_SPQ_MODE_EBLOCK) || - (found->queue == &p_spq->unlimited_pending)) + if (found->comp_mode != QED_SPQ_MODE_EBLOCK) /* EBLOCK is responsible for returning its own entry into the - * free list, unless it originally added the entry into the - * unlimited pending list. + * free list. */ qed_spq_return_entry(p_hwfn, found); -- cgit v1.2.3 From fb5e7438e7a3c8966e04ccb0760170e9e06f3699 Mon Sep 17 00:00:00 2001 From: Denis Bolotin Date: Thu, 8 Nov 2018 16:46:10 +0200 Subject: qed: Fix SPQ entries not returned to pool in error flows qed_sp_destroy_request() API was added for SPQ users that need to free/return the entry they acquired in their error flows. Signed-off-by: Denis Bolotin Signed-off-by: Michal Kalderon Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_fcoe.c | 11 ++++++++--- drivers/net/ethernet/qlogic/qed/qed_iscsi.c | 1 + drivers/net/ethernet/qlogic/qed/qed_l2.c | 12 ++++++++---- drivers/net/ethernet/qlogic/qed/qed_rdma.c | 1 + drivers/net/ethernet/qlogic/qed/qed_roce.c | 1 + drivers/net/ethernet/qlogic/qed/qed_sp.h | 11 +++++++++++ drivers/net/ethernet/qlogic/qed/qed_sp_commands.c | 22 ++++++++++++++-------- drivers/net/ethernet/qlogic/qed/qed_sriov.c | 1 + 8 files changed, 45 insertions(+), 15 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/qlogic/qed/qed_fcoe.c b/drivers/net/ethernet/qlogic/qed/qed_fcoe.c index cc1b373c0ace..46dc93d3b9b5 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_fcoe.c +++ b/drivers/net/ethernet/qlogic/qed/qed_fcoe.c @@ -147,7 +147,8 @@ qed_sp_fcoe_func_start(struct qed_hwfn *p_hwfn, "Cannot satisfy CQ amount. CQs requested %d, CQs available %d. Aborting function start\n", fcoe_pf_params->num_cqs, p_hwfn->hw_info.feat_num[QED_FCOE_CQ]); - return -EINVAL; + rc = -EINVAL; + goto err; } p_data->mtu = cpu_to_le16(fcoe_pf_params->mtu); @@ -156,14 +157,14 @@ qed_sp_fcoe_func_start(struct qed_hwfn *p_hwfn, rc = qed_cxt_acquire_cid(p_hwfn, PROTOCOLID_FCOE, &dummy_cid); if (rc) - return rc; + goto err; cxt_info.iid = dummy_cid; rc = qed_cxt_get_cid_info(p_hwfn, &cxt_info); if (rc) { DP_NOTICE(p_hwfn, "Cannot find context info for dummy cid=%d\n", dummy_cid); - return rc; + goto err; } p_cxt = cxt_info.p_cxt; SET_FIELD(p_cxt->tstorm_ag_context.flags3, @@ -240,6 +241,10 @@ qed_sp_fcoe_func_start(struct qed_hwfn *p_hwfn, rc = qed_spq_post(p_hwfn, p_ent, NULL); return rc; + +err: + qed_sp_destroy_request(p_hwfn, p_ent); + return rc; } static int diff --git a/drivers/net/ethernet/qlogic/qed/qed_iscsi.c b/drivers/net/ethernet/qlogic/qed/qed_iscsi.c index 1135387bd99d..4f8a685d1a55 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_iscsi.c +++ b/drivers/net/ethernet/qlogic/qed/qed_iscsi.c @@ -200,6 +200,7 @@ qed_sp_iscsi_func_start(struct qed_hwfn *p_hwfn, "Cannot satisfy CQ amount. Queues requested %d, CQs available %d. Aborting function start\n", p_params->num_queues, p_hwfn->hw_info.feat_num[QED_ISCSI_CQ]); + qed_sp_destroy_request(p_hwfn, p_ent); return -EINVAL; } diff --git a/drivers/net/ethernet/qlogic/qed/qed_l2.c b/drivers/net/ethernet/qlogic/qed/qed_l2.c index 82a1bd1f8a8c..67c02ea93906 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_l2.c +++ b/drivers/net/ethernet/qlogic/qed/qed_l2.c @@ -740,8 +740,7 @@ int qed_sp_vport_update(struct qed_hwfn *p_hwfn, rc = qed_sp_vport_update_rss(p_hwfn, p_ramrod, p_rss_params); if (rc) { - /* Return spq entry which is taken in qed_sp_init_request()*/ - qed_spq_return_entry(p_hwfn, p_ent); + qed_sp_destroy_request(p_hwfn, p_ent); return rc; } @@ -1355,6 +1354,7 @@ qed_filter_ucast_common(struct qed_hwfn *p_hwfn, DP_NOTICE(p_hwfn, "%d is not supported yet\n", p_filter_cmd->opcode); + qed_sp_destroy_request(p_hwfn, *pp_ent); return -EINVAL; } @@ -2056,13 +2056,13 @@ qed_configure_rfs_ntuple_filter(struct qed_hwfn *p_hwfn, } else { rc = qed_fw_vport(p_hwfn, p_params->vport_id, &abs_vport_id); if (rc) - return rc; + goto err; if (p_params->qid != QED_RFS_NTUPLE_QID_RSS) { rc = qed_fw_l2_queue(p_hwfn, p_params->qid, &abs_rx_q_id); if (rc) - return rc; + goto err; p_ramrod->rx_qid_valid = 1; p_ramrod->rx_qid = cpu_to_le16(abs_rx_q_id); @@ -2083,6 +2083,10 @@ qed_configure_rfs_ntuple_filter(struct qed_hwfn *p_hwfn, (u64)p_params->addr, p_params->length); return qed_spq_post(p_hwfn, p_ent, NULL); + +err: + qed_sp_destroy_request(p_hwfn, p_ent); + return rc; } int qed_get_rxq_coalesce(struct qed_hwfn *p_hwfn, diff --git a/drivers/net/ethernet/qlogic/qed/qed_rdma.c b/drivers/net/ethernet/qlogic/qed/qed_rdma.c index c71391b9c757..62113438c880 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_rdma.c +++ b/drivers/net/ethernet/qlogic/qed/qed_rdma.c @@ -1514,6 +1514,7 @@ qed_rdma_register_tid(void *rdma_cxt, default: rc = -EINVAL; DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "rc = %d\n", rc); + qed_sp_destroy_request(p_hwfn, p_ent); return rc; } SET_FIELD(p_ramrod->flags1, diff --git a/drivers/net/ethernet/qlogic/qed/qed_roce.c b/drivers/net/ethernet/qlogic/qed/qed_roce.c index f9167d1354bb..e49fada85410 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_roce.c +++ b/drivers/net/ethernet/qlogic/qed/qed_roce.c @@ -745,6 +745,7 @@ static int qed_roce_sp_destroy_qp_responder(struct qed_hwfn *p_hwfn, DP_NOTICE(p_hwfn, "qed destroy responder failed: cannot allocate memory (ramrod). rc = %d\n", rc); + qed_sp_destroy_request(p_hwfn, p_ent); return rc; } diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp.h b/drivers/net/ethernet/qlogic/qed/qed_sp.h index 04259df8a5c2..3157c0d99441 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sp.h +++ b/drivers/net/ethernet/qlogic/qed/qed_sp.h @@ -399,6 +399,17 @@ struct qed_sp_init_data { struct qed_spq_comp_cb *p_comp_data; }; +/** + * @brief Returns a SPQ entry to the pool / frees the entry if allocated. + * Should be called on in error flows after initializing the SPQ entry + * and before posting it. + * + * @param p_hwfn + * @param p_ent + */ +void qed_sp_destroy_request(struct qed_hwfn *p_hwfn, + struct qed_spq_entry *p_ent); + int qed_sp_init_request(struct qed_hwfn *p_hwfn, struct qed_spq_entry **pp_ent, u8 cmd, diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c index e86a1ea23613..888274fa208b 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c @@ -47,6 +47,19 @@ #include "qed_sp.h" #include "qed_sriov.h" +void qed_sp_destroy_request(struct qed_hwfn *p_hwfn, + struct qed_spq_entry *p_ent) +{ + /* qed_spq_get_entry() can either get an entry from the free_pool, + * or, if no entries are left, allocate a new entry and add it to + * the unlimited_pending list. + */ + if (p_ent->queue == &p_hwfn->p_spq->unlimited_pending) + kfree(p_ent); + else + qed_spq_return_entry(p_hwfn, p_ent); +} + int qed_sp_init_request(struct qed_hwfn *p_hwfn, struct qed_spq_entry **pp_ent, u8 cmd, u8 protocol, struct qed_sp_init_data *p_data) @@ -111,14 +124,7 @@ int qed_sp_init_request(struct qed_hwfn *p_hwfn, return 0; err: - /* qed_spq_get_entry() can either get an entry from the free_pool, - * or, if no entries are left, allocate a new entry and add it to - * the unlimited_pending list. - */ - if (p_ent->queue == &p_hwfn->p_spq->unlimited_pending) - kfree(p_ent); - else - qed_spq_return_entry(p_hwfn, p_ent); + qed_sp_destroy_request(p_hwfn, p_ent); return -EINVAL; } diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c index 9b08a9d9e151..ca6290fa0f30 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c @@ -101,6 +101,7 @@ static int qed_sp_vf_start(struct qed_hwfn *p_hwfn, struct qed_vf_info *p_vf) default: DP_NOTICE(p_hwfn, "Unknown VF personality %d\n", p_hwfn->hw_info.personality); + qed_sp_destroy_request(p_hwfn, p_ent); return -EINVAL; } -- cgit v1.2.3 From fa5c448d98f0df660bfcad3dd5facc027ef84cd3 Mon Sep 17 00:00:00 2001 From: Sagiv Ozeri Date: Thu, 8 Nov 2018 16:46:11 +0200 Subject: qed: Fix potential memory corruption A stuck ramrod should be deleted from the completion_pending list, otherwise it will be added again in the future and corrupt the list. Return error value to inform that ramrod is stuck and should be deleted. Signed-off-by: Sagiv Ozeri Signed-off-by: Denis Bolotin Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_spq.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/qlogic/qed/qed_spq.c b/drivers/net/ethernet/qlogic/qed/qed_spq.c index c1a81ec0524b..0a9c5bb0fa48 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_spq.c +++ b/drivers/net/ethernet/qlogic/qed/qed_spq.c @@ -142,6 +142,7 @@ static int qed_spq_block(struct qed_hwfn *p_hwfn, DP_INFO(p_hwfn, "Ramrod is stuck, requesting MCP drain\n"); rc = qed_mcp_drain(p_hwfn, p_ptt); + qed_ptt_release(p_hwfn, p_ptt); if (rc) { DP_NOTICE(p_hwfn, "MCP drain failed\n"); goto err; @@ -150,18 +151,15 @@ static int qed_spq_block(struct qed_hwfn *p_hwfn, /* Retry after drain */ rc = __qed_spq_block(p_hwfn, p_ent, p_fw_ret, true); if (!rc) - goto out; + return 0; comp_done = (struct qed_spq_comp_done *)p_ent->comp_cb.cookie; - if (comp_done->done == 1) + if (comp_done->done == 1) { if (p_fw_ret) *p_fw_ret = comp_done->fw_return_code; -out: - qed_ptt_release(p_hwfn, p_ptt); - return 0; - + return 0; + } err: - qed_ptt_release(p_hwfn, p_ptt); DP_NOTICE(p_hwfn, "Ramrod is stuck [CID %08x cmd %02x protocol %02x echo %04x]\n", le32_to_cpu(p_ent->elem.hdr.cid), -- cgit v1.2.3 From 8137b6ef0ce469154e5cf19f8e7fe04d9a72ac5e Mon Sep 17 00:00:00 2001 From: Thor Thayer Date: Thu, 8 Nov 2018 11:42:14 -0600 Subject: net: stmmac: Fix RX packet size > 8191 Ping problems with packets > 8191 as shown: PING 192.168.1.99 (192.168.1.99) 8150(8178) bytes of data. 8158 bytes from 192.168.1.99: icmp_seq=1 ttl=64 time=0.669 ms wrong data byte 8144 should be 0xd0 but was 0x0 16 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f 20 21 22 23 24 25 26 27 28 29 2a 2b 2c 2d 2e 2f %< ---------------snip-------------------------------------- 8112 b0 b1 b2 b3 b4 b5 b6 b7 b8 b9 ba bb bc bd be bf c0 c1 c2 c3 c4 c5 c6 c7 c8 c9 ca cb cc cd ce cf 8144 0 0 0 0 d0 d1 ^^^^^^^ Notice the 4 bytes of 0 before the expected byte of d0. Databook notes that the RX buffer must be a multiple of 4/8/16 bytes [1]. Update the DMA Buffer size define to 8188 instead of 8192. Remove the -1 from the RX buffer size allocations and use the new DMA Buffer size directly. [1] Synopsys DesignWare Cores Ethernet MAC Universal v3.70a [section 8.4.2 - Table 8-24] Tested on SoCFPGA Stratix10 with ping sweep from 100 to 8300 byte packets. Fixes: 286a83721720 ("stmmac: add CHAINED descriptor mode support (V4)") Suggested-by: Jose Abreu Signed-off-by: Thor Thayer Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/common.h | 3 ++- drivers/net/ethernet/stmicro/stmmac/descs_com.h | 2 +- drivers/net/ethernet/stmicro/stmmac/enh_desc.c | 2 +- drivers/net/ethernet/stmicro/stmmac/ring_mode.c | 2 +- 4 files changed, 5 insertions(+), 4 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index b1b305f8f414..272b9ca66314 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -365,7 +365,8 @@ struct dma_features { /* GMAC TX FIFO is 8K, Rx FIFO is 16K */ #define BUF_SIZE_16KiB 16384 -#define BUF_SIZE_8KiB 8192 +/* RX Buffer size must be < 8191 and multiple of 4/8/16 bytes */ +#define BUF_SIZE_8KiB 8188 #define BUF_SIZE_4KiB 4096 #define BUF_SIZE_2KiB 2048 diff --git a/drivers/net/ethernet/stmicro/stmmac/descs_com.h b/drivers/net/ethernet/stmicro/stmmac/descs_com.h index ca9d7e48034c..40d6356a7e73 100644 --- a/drivers/net/ethernet/stmicro/stmmac/descs_com.h +++ b/drivers/net/ethernet/stmicro/stmmac/descs_com.h @@ -31,7 +31,7 @@ /* Enhanced descriptors */ static inline void ehn_desc_rx_set_on_ring(struct dma_desc *p, int end) { - p->des1 |= cpu_to_le32(((BUF_SIZE_8KiB - 1) + p->des1 |= cpu_to_le32((BUF_SIZE_8KiB << ERDES1_BUFFER2_SIZE_SHIFT) & ERDES1_BUFFER2_SIZE_MASK); diff --git a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c index 77914c89d749..5ef91a790f9d 100644 --- a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c +++ b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c @@ -262,7 +262,7 @@ static void enh_desc_init_rx_desc(struct dma_desc *p, int disable_rx_ic, int mode, int end) { p->des0 |= cpu_to_le32(RDES0_OWN); - p->des1 |= cpu_to_le32((BUF_SIZE_8KiB - 1) & ERDES1_BUFFER1_SIZE_MASK); + p->des1 |= cpu_to_le32(BUF_SIZE_8KiB & ERDES1_BUFFER1_SIZE_MASK); if (mode == STMMAC_CHAIN_MODE) ehn_desc_rx_set_on_chain(p); diff --git a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c index abc3f85270cd..d8c5bc412219 100644 --- a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c +++ b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c @@ -140,7 +140,7 @@ static void clean_desc3(void *priv_ptr, struct dma_desc *p) static int set_16kib_bfsize(int mtu) { int ret = 0; - if (unlikely(mtu >= BUF_SIZE_8KiB)) + if (unlikely(mtu > BUF_SIZE_8KiB)) ret = BUF_SIZE_16KiB; return ret; } -- cgit v1.2.3 From 35e8e8b45d31bec34379dd36e7b71448e003efb2 Mon Sep 17 00:00:00 2001 From: Igor Russkikh Date: Fri, 9 Nov 2018 11:53:56 +0000 Subject: net: aquantia: synchronized flow control between mac/phy Flow control statuses were not synchronized between blocks, that caused packets/link drop on some corner cases, when MAC sent PFC although Phy was not expecting these to come. Driver should readout the negotiated FC from phy and configure RX block accordigly. This is done on each link change event with information from FW. Fixes: 288551de45aa ("net: aquantia: Implement rx/tx flow control ethtools callback") Signed-off-by: Igor Russkikh Signed-off-by: David S. Miller --- drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c | 8 ++++---- drivers/net/ethernet/aquantia/atlantic/aq_hw.h | 3 +++ drivers/net/ethernet/aquantia/atlantic/aq_nic.c | 14 +++++++++++++- .../ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c | 12 +++++++++--- .../aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c | 21 +++++++++++++++++++++ 5 files changed, 50 insertions(+), 8 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c index 6a633c70f603..99ef1daaa4d8 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c @@ -407,13 +407,13 @@ static void aq_ethtool_get_pauseparam(struct net_device *ndev, struct ethtool_pauseparam *pause) { struct aq_nic_s *aq_nic = netdev_priv(ndev); + u32 fc = aq_nic->aq_nic_cfg.flow_control; pause->autoneg = 0; - if (aq_nic->aq_hw->aq_nic_cfg->flow_control & AQ_NIC_FC_RX) - pause->rx_pause = 1; - if (aq_nic->aq_hw->aq_nic_cfg->flow_control & AQ_NIC_FC_TX) - pause->tx_pause = 1; + pause->rx_pause = !!(fc & AQ_NIC_FC_RX); + pause->tx_pause = !!(fc & AQ_NIC_FC_TX); + } static int aq_ethtool_set_pauseparam(struct net_device *ndev, diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h index e8689241204e..7ec8d24b2b0b 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h @@ -204,6 +204,7 @@ struct aq_hw_ops { int (*hw_get_fw_version)(struct aq_hw_s *self, u32 *fw_version); + int (*hw_set_fc)(struct aq_hw_s *self, u32 fc, u32 tc); }; struct aq_fw_ops { @@ -226,6 +227,8 @@ struct aq_fw_ops { int (*update_stats)(struct aq_hw_s *self); + u32 (*get_flow_control)(struct aq_hw_s *self, u32 *fcmode); + int (*set_flow_control)(struct aq_hw_s *self); int (*set_power)(struct aq_hw_s *self, unsigned int power_state, diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c index 5fed24446687..0011a3f2f672 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c @@ -124,6 +124,7 @@ void aq_nic_cfg_start(struct aq_nic_s *self) static int aq_nic_update_link_status(struct aq_nic_s *self) { int err = self->aq_fw_ops->update_link_status(self->aq_hw); + u32 fc = 0; if (err) return err; @@ -133,6 +134,15 @@ static int aq_nic_update_link_status(struct aq_nic_s *self) AQ_CFG_DRV_NAME, self->link_status.mbps, self->aq_hw->aq_link_status.mbps); aq_nic_update_interrupt_moderation_settings(self); + + /* Driver has to update flow control settings on RX block + * on any link event. + * We should query FW whether it negotiated FC. + */ + if (self->aq_fw_ops->get_flow_control) + self->aq_fw_ops->get_flow_control(self->aq_hw, &fc); + if (self->aq_hw_ops->hw_set_fc) + self->aq_hw_ops->hw_set_fc(self->aq_hw, fc, 0); } self->link_status = self->aq_hw->aq_link_status; @@ -772,7 +782,9 @@ void aq_nic_get_link_ksettings(struct aq_nic_s *self, ethtool_link_ksettings_add_link_mode(cmd, advertising, Pause); - if (self->aq_nic_cfg.flow_control & AQ_NIC_FC_TX) + /* Asym is when either RX or TX, but not both */ + if (!!(self->aq_nic_cfg.flow_control & AQ_NIC_FC_TX) ^ + !!(self->aq_nic_cfg.flow_control & AQ_NIC_FC_RX)) ethtool_link_ksettings_add_link_mode(cmd, advertising, Asym_Pause); diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c index 76d25d594a0f..119265762b0c 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c @@ -100,12 +100,17 @@ static int hw_atl_b0_hw_reset(struct aq_hw_s *self) return err; } +static int hw_atl_b0_set_fc(struct aq_hw_s *self, u32 fc, u32 tc) +{ + hw_atl_rpb_rx_xoff_en_per_tc_set(self, !!(fc & AQ_NIC_FC_RX), tc); + return 0; +} + static int hw_atl_b0_hw_qos_set(struct aq_hw_s *self) { u32 tc = 0U; u32 buff_size = 0U; unsigned int i_priority = 0U; - bool is_rx_flow_control = false; /* TPS Descriptor rate init */ hw_atl_tps_tx_pkt_shed_desc_rate_curr_time_res_set(self, 0x0U); @@ -138,7 +143,6 @@ static int hw_atl_b0_hw_qos_set(struct aq_hw_s *self) /* QoS Rx buf size per TC */ tc = 0; - is_rx_flow_control = (AQ_NIC_FC_RX & self->aq_nic_cfg->flow_control); buff_size = HW_ATL_B0_RXBUF_MAX; hw_atl_rpb_rx_pkt_buff_size_per_tc_set(self, buff_size, tc); @@ -150,7 +154,8 @@ static int hw_atl_b0_hw_qos_set(struct aq_hw_s *self) (buff_size * (1024U / 32U) * 50U) / 100U, tc); - hw_atl_rpb_rx_xoff_en_per_tc_set(self, is_rx_flow_control ? 1U : 0U, tc); + + hw_atl_b0_set_fc(self, self->aq_nic_cfg->flow_control, tc); /* QoS 802.1p priority -> TC mapping */ for (i_priority = 8U; i_priority--;) @@ -963,4 +968,5 @@ const struct aq_hw_ops hw_atl_ops_b0 = { .hw_get_regs = hw_atl_utils_hw_get_regs, .hw_get_hw_stats = hw_atl_utils_get_hw_stats, .hw_get_fw_version = hw_atl_utils_get_fw_version, + .hw_set_fc = hw_atl_b0_set_fc, }; diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c index 096ca5730887..7de3220d9cab 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c @@ -30,6 +30,8 @@ #define HW_ATL_FW2X_MPI_STATE_ADDR 0x370 #define HW_ATL_FW2X_MPI_STATE2_ADDR 0x374 +#define HW_ATL_FW2X_CAP_PAUSE BIT(CAPS_HI_PAUSE) +#define HW_ATL_FW2X_CAP_ASYM_PAUSE BIT(CAPS_HI_ASYMMETRIC_PAUSE) #define HW_ATL_FW2X_CAP_SLEEP_PROXY BIT(CAPS_HI_SLEEP_PROXY) #define HW_ATL_FW2X_CAP_WOL BIT(CAPS_HI_WOL) @@ -451,6 +453,24 @@ static int aq_fw2x_set_flow_control(struct aq_hw_s *self) return 0; } +static u32 aq_fw2x_get_flow_control(struct aq_hw_s *self, u32 *fcmode) +{ + u32 mpi_state = aq_hw_read_reg(self, HW_ATL_FW2X_MPI_STATE2_ADDR); + + if (mpi_state & HW_ATL_FW2X_CAP_PAUSE) + if (mpi_state & HW_ATL_FW2X_CAP_ASYM_PAUSE) + *fcmode = AQ_NIC_FC_RX; + else + *fcmode = AQ_NIC_FC_RX | AQ_NIC_FC_TX; + else + if (mpi_state & HW_ATL_FW2X_CAP_ASYM_PAUSE) + *fcmode = AQ_NIC_FC_TX; + else + *fcmode = 0; + + return 0; +} + const struct aq_fw_ops aq_fw_2x_ops = { .init = aq_fw2x_init, .deinit = aq_fw2x_deinit, @@ -465,4 +485,5 @@ const struct aq_fw_ops aq_fw_2x_ops = { .set_eee_rate = aq_fw2x_set_eee_rate, .get_eee_rate = aq_fw2x_get_eee_rate, .set_flow_control = aq_fw2x_set_flow_control, + .get_flow_control = aq_fw2x_get_flow_control }; -- cgit v1.2.3 From 7a1bb49461b12b2e6332a4d054256835f45203f3 Mon Sep 17 00:00:00 2001 From: Dmitry Bogdanov Date: Fri, 9 Nov 2018 11:53:57 +0000 Subject: net: aquantia: fix potential IOMMU fault after driver unbind IOMMU fault may occurr on unbind/bind or if_down/if_up sequence. Although driver disables the rings on down, this is not enough. Due to internal HW design, during subsequent initialization NIC sometimes may reuse RX descriptors cache and write to the host memory from the descriptor cache. That's get catched by IOMMU on host. This patch invalidates the descriptor cache in NIC on interface down to prevent writing to the cached descriptors and to the memory pointed in those descriptors. Signed-off-by: Dmitry Bogdanov Signed-off-by: Igor Russkikh Signed-off-by: David S. Miller --- .../net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c | 6 ++++++ .../net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c | 8 ++++++++ .../net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h | 3 +++ .../aquantia/atlantic/hw_atl/hw_atl_llh_internal.h | 18 ++++++++++++++++++ 4 files changed, 35 insertions(+) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c index 119265762b0c..3aec56623bf5 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c @@ -920,6 +920,12 @@ static int hw_atl_b0_hw_interrupt_moderation_set(struct aq_hw_s *self) static int hw_atl_b0_hw_stop(struct aq_hw_s *self) { hw_atl_b0_hw_irq_disable(self, HW_ATL_B0_INT_MASK); + + /* Invalidate Descriptor Cache to prevent writing to the cached + * descriptors and to the data pointer of those descriptors + */ + hw_atl_rdm_rx_dma_desc_cache_init_set(self, 1); + return aq_hw_err_from_flags(self); } diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c index be0a3a90dfad..5502ec5f0f69 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c @@ -619,6 +619,14 @@ void hw_atl_rpb_rx_flow_ctl_mode_set(struct aq_hw_s *aq_hw, u32 rx_flow_ctl_mode HW_ATL_RPB_RX_FC_MODE_SHIFT, rx_flow_ctl_mode); } +void hw_atl_rdm_rx_dma_desc_cache_init_set(struct aq_hw_s *aq_hw, u32 init) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_ADR, + HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_MSK, + HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_SHIFT, + init); +} + void hw_atl_rpb_rx_pkt_buff_size_per_tc_set(struct aq_hw_s *aq_hw, u32 rx_pkt_buff_size_per_tc, u32 buffer) { diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h index 7056c7342afc..41f239928c15 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h @@ -325,6 +325,9 @@ void hw_atl_rpb_rx_pkt_buff_size_per_tc_set(struct aq_hw_s *aq_hw, u32 rx_pkt_buff_size_per_tc, u32 buffer); +/* set rdm rx dma descriptor cache init */ +void hw_atl_rdm_rx_dma_desc_cache_init_set(struct aq_hw_s *aq_hw, u32 init); + /* set rx xoff enable (per tc) */ void hw_atl_rpb_rx_xoff_en_per_tc_set(struct aq_hw_s *aq_hw, u32 rx_xoff_en_per_tc, u32 buffer); diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h index 716674a9b729..a715fa317b1c 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h @@ -293,6 +293,24 @@ /* default value of bitfield desc{d}_reset */ #define HW_ATL_RDM_DESCDRESET_DEFAULT 0x0 +/* rdm_desc_init_i bitfield definitions + * preprocessor definitions for the bitfield rdm_desc_init_i. + * port="pif_rdm_desc_init_i" + */ + +/* register address for bitfield rdm_desc_init_i */ +#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_ADR 0x00005a00 +/* bitmask for bitfield rdm_desc_init_i */ +#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_MSK 0xffffffff +/* inverted bitmask for bitfield rdm_desc_init_i */ +#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_MSKN 0x00000000 +/* lower bit position of bitfield rdm_desc_init_i */ +#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_SHIFT 0 +/* width of bitfield rdm_desc_init_i */ +#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_WIDTH 32 +/* default value of bitfield rdm_desc_init_i */ +#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_DEFAULT 0x0 + /* rx int_desc_wrb_en bitfield definitions * preprocessor definitions for the bitfield "int_desc_wrb_en". * port="pif_rdm_int_desc_wrb_en_i" -- cgit v1.2.3 From bfaa9f8553d5c20703781e63f4fc8cb4792f18fd Mon Sep 17 00:00:00 2001 From: Igor Russkikh Date: Fri, 9 Nov 2018 11:53:59 +0000 Subject: net: aquantia: fixed enable unicast on 32 macvlan Fixed a condition mistake due to which macvlans unicast item number 32 was not added in the unicast filter. The consequence is that when exactly 32 macvlans are created on NIC, the last created macvlan receives no traffic because its MAC was not registered in HW. Fixes: 94b3b542303f ("net: aquantia: vlan unicast address list correct handling") Signed-off-by: Igor Russkikh Tested-by: Nikita Danilov Signed-off-by: David S. Miller --- drivers/net/ethernet/aquantia/atlantic/aq_nic.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c index 0011a3f2f672..b5e7c98f424c 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c @@ -600,7 +600,7 @@ int aq_nic_set_multicast_list(struct aq_nic_s *self, struct net_device *ndev) } } - if (i > 0 && i < AQ_HW_MULTICAST_ADDRESS_MAX) { + if (i > 0 && i <= AQ_HW_MULTICAST_ADDRESS_MAX) { packet_filter |= IFF_MULTICAST; self->mc_list.count = i; self->aq_hw_ops->hw_multicast_list_set(self->aq_hw, -- cgit v1.2.3 From ad703c2b9127f9acdef697ec4755f6da4beaa266 Mon Sep 17 00:00:00 2001 From: Dmitry Bogdanov Date: Fri, 9 Nov 2018 11:54:01 +0000 Subject: net: aquantia: invalid checksumm offload implementation Packets with marked invalid IP/UDP/TCP checksums were considered as good by the driver. The error was in a logic, processing offload bits in RX descriptor. Signed-off-by: Igor Russkikh Signed-off-by: Dmitry Bogdanov Signed-off-by: David S. Miller --- drivers/net/ethernet/aquantia/atlantic/aq_ring.c | 35 +++++++++++++-------- .../ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c | 36 +++++++++++----------- 2 files changed, 41 insertions(+), 30 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ring.c b/drivers/net/ethernet/aquantia/atlantic/aq_ring.c index 3db91446cc67..74550ccc7a20 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_ring.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_ring.c @@ -172,6 +172,27 @@ bool aq_ring_tx_clean(struct aq_ring_s *self) return !!budget; } +static void aq_rx_checksum(struct aq_ring_s *self, + struct aq_ring_buff_s *buff, + struct sk_buff *skb) +{ + if (!(self->aq_nic->ndev->features & NETIF_F_RXCSUM)) + return; + + if (unlikely(buff->is_cso_err)) { + ++self->stats.rx.errors; + skb->ip_summed = CHECKSUM_NONE; + return; + } + if (buff->is_ip_cso) { + __skb_incr_checksum_unnecessary(skb); + if (buff->is_udp_cso || buff->is_tcp_cso) + __skb_incr_checksum_unnecessary(skb); + } else { + skb->ip_summed = CHECKSUM_NONE; + } +} + #define AQ_SKB_ALIGN SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) int aq_ring_rx_clean(struct aq_ring_s *self, struct napi_struct *napi, @@ -267,18 +288,8 @@ int aq_ring_rx_clean(struct aq_ring_s *self, } skb->protocol = eth_type_trans(skb, ndev); - if (unlikely(buff->is_cso_err)) { - ++self->stats.rx.errors; - skb->ip_summed = CHECKSUM_NONE; - } else { - if (buff->is_ip_cso) { - __skb_incr_checksum_unnecessary(skb); - if (buff->is_udp_cso || buff->is_tcp_cso) - __skb_incr_checksum_unnecessary(skb); - } else { - skb->ip_summed = CHECKSUM_NONE; - } - } + + aq_rx_checksum(self, buff, skb); skb_set_hash(skb, buff->rss_hash, buff->is_hash_l4 ? PKT_HASH_TYPE_L4 : diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c index 3aec56623bf5..179ce12fe4d8 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c @@ -660,9 +660,9 @@ static int hw_atl_b0_hw_ring_rx_receive(struct aq_hw_s *self, struct hw_atl_rxd_wb_s *rxd_wb = (struct hw_atl_rxd_wb_s *) &ring->dx_ring[ring->hw_head * HW_ATL_B0_RXD_SIZE]; - unsigned int is_err = 1U; unsigned int is_rx_check_sum_enabled = 0U; unsigned int pkt_type = 0U; + u8 rx_stat = 0U; if (!(rxd_wb->status & 0x1U)) { /* RxD is not done */ break; @@ -670,35 +670,35 @@ static int hw_atl_b0_hw_ring_rx_receive(struct aq_hw_s *self, buff = &ring->buff_ring[ring->hw_head]; - is_err = (0x0000003CU & rxd_wb->status); + rx_stat = (0x0000003CU & rxd_wb->status) >> 2; is_rx_check_sum_enabled = (rxd_wb->type) & (0x3U << 19); - is_err &= ~0x20U; /* exclude validity bit */ pkt_type = 0xFFU & (rxd_wb->type >> 4); - if (is_rx_check_sum_enabled) { - if (0x0U == (pkt_type & 0x3U)) - buff->is_ip_cso = (is_err & 0x08U) ? 0U : 1U; + if (is_rx_check_sum_enabled & BIT(0) && + (0x0U == (pkt_type & 0x3U))) + buff->is_ip_cso = (rx_stat & BIT(1)) ? 0U : 1U; + if (is_rx_check_sum_enabled & BIT(1)) { if (0x4U == (pkt_type & 0x1CU)) - buff->is_udp_cso = buff->is_cso_err ? 0U : 1U; + buff->is_udp_cso = (rx_stat & BIT(2)) ? 0U : + !!(rx_stat & BIT(3)); else if (0x0U == (pkt_type & 0x1CU)) - buff->is_tcp_cso = buff->is_cso_err ? 0U : 1U; - - /* Checksum offload workaround for small packets */ - if (rxd_wb->pkt_len <= 60) { - buff->is_ip_cso = 0U; - buff->is_cso_err = 0U; - } + buff->is_tcp_cso = (rx_stat & BIT(2)) ? 0U : + !!(rx_stat & BIT(3)); + } + buff->is_cso_err = !!(rx_stat & 0x6); + /* Checksum offload workaround for small packets */ + if (unlikely(rxd_wb->pkt_len <= 60)) { + buff->is_ip_cso = 0U; + buff->is_cso_err = 0U; } - - is_err &= ~0x18U; dma_unmap_page(ndev, buff->pa, buff->len, DMA_FROM_DEVICE); - if (is_err || rxd_wb->type & 0x1000U) { - /* status error or DMA error */ + if ((rx_stat & BIT(0)) || rxd_wb->type & 0x1000U) { + /* MAC error or DMA error */ buff->is_error = 1U; } else { if (self->aq_nic_cfg->is_rss) { -- cgit v1.2.3 From bbb67a44baf973da734b9fd61cba4211da240751 Mon Sep 17 00:00:00 2001 From: Dmitry Bogdanov Date: Fri, 9 Nov 2018 11:54:03 +0000 Subject: net: aquantia: allow rx checksum offload configuration RX Checksum offloads could not be configured and ignored netdev features flag for checksumming. Signed-off-by: Igor Russkikh Signed-off-by: Dmitry Bogdanov Signed-off-by: David S. Miller --- drivers/net/ethernet/aquantia/atlantic/aq_hw.h | 3 +++ drivers/net/ethernet/aquantia/atlantic/aq_main.c | 10 ++++++++-- drivers/net/ethernet/aquantia/atlantic/aq_nic.c | 2 +- drivers/net/ethernet/aquantia/atlantic/aq_nic.h | 2 +- drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c | 7 +++++-- 5 files changed, 18 insertions(+), 6 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h index 7ec8d24b2b0b..a1e70da358ca 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h @@ -204,6 +204,9 @@ struct aq_hw_ops { int (*hw_get_fw_version)(struct aq_hw_s *self, u32 *fw_version); + int (*hw_set_offload)(struct aq_hw_s *self, + struct aq_nic_cfg_s *aq_nic_cfg); + int (*hw_set_fc)(struct aq_hw_s *self, u32 fc, u32 tc); }; diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_main.c b/drivers/net/ethernet/aquantia/atlantic/aq_main.c index e3ae29e523f0..7c07eef275eb 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_main.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_main.c @@ -99,8 +99,11 @@ static int aq_ndev_set_features(struct net_device *ndev, struct aq_nic_s *aq_nic = netdev_priv(ndev); struct aq_nic_cfg_s *aq_cfg = aq_nic_get_cfg(aq_nic); bool is_lro = false; + int err = 0; + + aq_cfg->features = features; - if (aq_cfg->hw_features & NETIF_F_LRO) { + if (aq_cfg->aq_hw_caps->hw_features & NETIF_F_LRO) { is_lro = features & NETIF_F_LRO; if (aq_cfg->is_lro != is_lro) { @@ -112,8 +115,11 @@ static int aq_ndev_set_features(struct net_device *ndev, } } } + if ((aq_nic->ndev->features ^ features) & NETIF_F_RXCSUM) + err = aq_nic->aq_hw_ops->hw_set_offload(aq_nic->aq_hw, + aq_cfg); - return 0; + return err; } static int aq_ndev_set_mac_address(struct net_device *ndev, void *addr) diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c index b5e7c98f424c..7abdc0952425 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c @@ -118,7 +118,7 @@ void aq_nic_cfg_start(struct aq_nic_s *self) } cfg->link_speed_msk &= cfg->aq_hw_caps->link_speed_msk; - cfg->hw_features = cfg->aq_hw_caps->hw_features; + cfg->features = cfg->aq_hw_caps->hw_features; } static int aq_nic_update_link_status(struct aq_nic_s *self) diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.h b/drivers/net/ethernet/aquantia/atlantic/aq_nic.h index c1582f4e8e1b..44ec47a3d60a 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.h @@ -23,7 +23,7 @@ struct aq_vec_s; struct aq_nic_cfg_s { const struct aq_hw_caps_s *aq_hw_caps; - u64 hw_features; + u64 features; u32 rxds; /* rx ring size, descriptors # */ u32 txds; /* tx ring size, descriptors # */ u32 vecs; /* vecs==allocated irqs */ diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c index 179ce12fe4d8..f02592f43fe3 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c @@ -234,8 +234,10 @@ static int hw_atl_b0_hw_offload_set(struct aq_hw_s *self, hw_atl_tpo_tcp_udp_crc_offload_en_set(self, 1); /* RX checksums offloads*/ - hw_atl_rpo_ipv4header_crc_offload_en_set(self, 1); - hw_atl_rpo_tcp_udp_crc_offload_en_set(self, 1); + hw_atl_rpo_ipv4header_crc_offload_en_set(self, !!(aq_nic_cfg->features & + NETIF_F_RXCSUM)); + hw_atl_rpo_tcp_udp_crc_offload_en_set(self, !!(aq_nic_cfg->features & + NETIF_F_RXCSUM)); /* LSO offloads*/ hw_atl_tdm_large_send_offload_en_set(self, 0xFFFFFFFFU); @@ -974,5 +976,6 @@ const struct aq_hw_ops hw_atl_ops_b0 = { .hw_get_regs = hw_atl_utils_hw_get_regs, .hw_get_hw_stats = hw_atl_utils_get_hw_stats, .hw_get_fw_version = hw_atl_utils_get_fw_version, + .hw_set_offload = hw_atl_b0_hw_offload_set, .hw_set_fc = hw_atl_b0_set_fc, }; -- cgit v1.2.3 From d02854dc1999ed3e7fd79ec700c64ac23ac0c458 Mon Sep 17 00:00:00 2001 From: Subash Abhinov Kasiviswanathan Date: Fri, 9 Nov 2018 18:56:27 -0700 Subject: net: qualcomm: rmnet: Fix incorrect assignment of real_dev A null dereference was observed when a sysctl was being set from userspace and rmnet was stuck trying to complete some actions in the NETDEV_REGISTER callback. This is because the real_dev is set only after the device registration handler completes. sysctl call stack - <6> Unable to handle kernel NULL pointer dereference at virtual address 00000108 <2> pc : rmnet_vnd_get_iflink+0x1c/0x28 <2> lr : dev_get_iflink+0x2c/0x40 <2> rmnet_vnd_get_iflink+0x1c/0x28 <2> inet6_fill_ifinfo+0x15c/0x234 <2> inet6_ifinfo_notify+0x68/0xd4 <2> ndisc_ifinfo_sysctl_change+0x1b8/0x234 <2> proc_sys_call_handler+0xac/0x100 <2> proc_sys_write+0x3c/0x4c <2> __vfs_write+0x54/0x14c <2> vfs_write+0xcc/0x188 <2> SyS_write+0x60/0xc0 <2> el0_svc_naked+0x34/0x38 device register call stack - <2> notifier_call_chain+0x84/0xbc <2> raw_notifier_call_chain+0x38/0x48 <2> call_netdevice_notifiers_info+0x40/0x70 <2> call_netdevice_notifiers+0x38/0x60 <2> register_netdevice+0x29c/0x3d8 <2> rmnet_vnd_newlink+0x68/0xe8 <2> rmnet_newlink+0xa0/0x160 <2> rtnl_newlink+0x57c/0x6c8 <2> rtnetlink_rcv_msg+0x1dc/0x328 <2> netlink_rcv_skb+0xac/0x118 <2> rtnetlink_rcv+0x24/0x30 <2> netlink_unicast+0x158/0x1f0 <2> netlink_sendmsg+0x32c/0x338 <2> sock_sendmsg+0x44/0x60 <2> SyS_sendto+0x150/0x1ac <2> el0_svc_naked+0x34/0x38 Fixes: b752eff5be24 ("net: qualcomm: rmnet: Implement ndo_get_iflink") Signed-off-by: Sean Tranchetti Signed-off-by: Subash Abhinov Kasiviswanathan Signed-off-by: David S. Miller --- drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c index 0afc3d335d56..d11c16aeb19a 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c @@ -234,7 +234,7 @@ int rmnet_vnd_newlink(u8 id, struct net_device *rmnet_dev, struct net_device *real_dev, struct rmnet_endpoint *ep) { - struct rmnet_priv *priv; + struct rmnet_priv *priv = netdev_priv(rmnet_dev); int rc; if (ep->egress_dev) @@ -247,6 +247,8 @@ int rmnet_vnd_newlink(u8 id, struct net_device *rmnet_dev, rmnet_dev->hw_features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM; rmnet_dev->hw_features |= NETIF_F_SG; + priv->real_dev = real_dev; + rc = register_netdevice(rmnet_dev); if (!rc) { ep->egress_dev = rmnet_dev; @@ -255,9 +257,7 @@ int rmnet_vnd_newlink(u8 id, struct net_device *rmnet_dev, rmnet_dev->rtnl_link_ops = &rmnet_link_ops; - priv = netdev_priv(rmnet_dev); priv->mux_id = id; - priv->real_dev = real_dev; netdev_dbg(rmnet_dev, "rmnet dev created\n"); } -- cgit v1.2.3 From fbd1d5245372e48b494120a30fe0b34b304576c4 Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Fri, 9 Nov 2018 17:37:20 +0100 Subject: net: mvneta: correct typo The reserved variable should be named reserved1. Signed-off-by: Alexandre Belloni Signed-off-by: David S. Miller --- drivers/net/ethernet/marvell/mvneta.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c index 5bfd349bf41a..3ba672e9e353 100644 --- a/drivers/net/ethernet/marvell/mvneta.c +++ b/drivers/net/ethernet/marvell/mvneta.c @@ -494,7 +494,7 @@ struct mvneta_port { #if defined(__LITTLE_ENDIAN) struct mvneta_tx_desc { u32 command; /* Options used by HW for packet transmitting.*/ - u16 reserverd1; /* csum_l4 (for future use) */ + u16 reserved1; /* csum_l4 (for future use) */ u16 data_size; /* Data size of transmitted packet in bytes */ u32 buf_phys_addr; /* Physical addr of transmitted buffer */ u32 reserved2; /* hw_cmd - (for future use, PMT) */ @@ -519,7 +519,7 @@ struct mvneta_rx_desc { #else struct mvneta_tx_desc { u16 data_size; /* Data size of transmitted packet in bytes */ - u16 reserverd1; /* csum_l4 (for future use) */ + u16 reserved1; /* csum_l4 (for future use) */ u32 command; /* Options used by HW for packet transmitting.*/ u32 reserved2; /* hw_cmd - (for future use, PMT) */ u32 buf_phys_addr; /* Physical addr of transmitted buffer */ -- cgit v1.2.3 From 77e461d14ed141253573eeeb4d34eccc51e38328 Mon Sep 17 00:00:00 2001 From: Sudarsana Reddy Kalluru Date: Sun, 11 Nov 2018 18:27:34 -0800 Subject: bnx2x: Assign unique DMAE channel number for FW DMAE transactions. Driver assigns DMAE channel 0 for FW as part of START_RAMROD command. FW uses this channel for DMAE operations (e.g., TIME_SYNC implementation). Driver also uses the same channel 0 for DMAE operations for some of the PFs (e.g., PF0 on Port0). This could lead to concurrent access to the DMAE channel by FW and driver which is not legal. Hence need to assign unique DMAE id for FW. Currently following DMAE channels are used by the clients, MFW - OCBB/OCSD functionality uses DMAE channel 14/15 Driver 0-3 and 8-11 (for PF dmae operations) 4 and 12 (for stats requests) Assigning unique dmae_id '13' to the FW. Changes from previous version: ------------------------------ v2: Incorporated the review comments. Signed-off-by: Sudarsana Reddy Kalluru Signed-off-by: Michal Kalderon Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x.h | 7 +++++++ drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c | 1 + 2 files changed, 8 insertions(+) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h index be1506169076..0de487a8f0eb 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h @@ -2191,6 +2191,13 @@ void bnx2x_igu_clear_sb_gen(struct bnx2x *bp, u8 func, u8 idu_sb_id, #define PMF_DMAE_C(bp) (BP_PORT(bp) * MAX_DMAE_C_PER_PORT + \ E1HVN_MAX) +/* Following is the DMAE channel number allocation for the clients. + * MFW: OCBB/OCSD implementations use DMAE channels 14/15 respectively. + * Driver: 0-3 and 8-11 (for PF dmae operations) + * 4 and 12 (for stats requests) + */ +#define BNX2X_FW_DMAE_C 13 /* Channel for FW DMAE operations */ + /* PCIE link and speed */ #define PCICFG_LINK_WIDTH 0x1f00000 #define PCICFG_LINK_WIDTH_SHIFT 20 diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c index 3f4d2c8da21a..a9eaaf3e73a4 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c @@ -6149,6 +6149,7 @@ static inline int bnx2x_func_send_start(struct bnx2x *bp, rdata->sd_vlan_tag = cpu_to_le16(start_params->sd_vlan_tag); rdata->path_id = BP_PATH(bp); rdata->network_cos_mode = start_params->network_cos_mode; + rdata->dmae_cmd_id = BNX2X_FW_DMAE_C; rdata->vxlan_dst_port = cpu_to_le16(start_params->vxlan_dst_port); rdata->geneve_dst_port = cpu_to_le16(start_params->geneve_dst_port); -- cgit v1.2.3 From 9aaa4e8ba12972d674caeefbc5f88d83235dd697 Mon Sep 17 00:00:00 2001 From: Denis Bolotin Date: Mon, 12 Nov 2018 12:50:20 +0200 Subject: qed: Fix PTT leak in qed_drain() Release PTT before entering error flow. Signed-off-by: Denis Bolotin Signed-off-by: Michal Kalderon Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c index 35fd0db6a677..fff7f04d4525 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_main.c +++ b/drivers/net/ethernet/qlogic/qed/qed_main.c @@ -1782,9 +1782,9 @@ static int qed_drain(struct qed_dev *cdev) return -EBUSY; } rc = qed_mcp_drain(hwfn, ptt); + qed_ptt_release(hwfn, ptt); if (rc) return rc; - qed_ptt_release(hwfn, ptt); } return 0; -- cgit v1.2.3 From e90202ed1cf9672c48a363c84a929932ebfe6fc0 Mon Sep 17 00:00:00 2001 From: Denis Bolotin Date: Mon, 12 Nov 2018 12:50:21 +0200 Subject: qed: Fix overriding offload_tc by protocols without APP TLV The TC received from APP TLV is stored in offload_tc, and should not be set by protocols which did not receive an APP TLV. Fixed the condition when overriding the offload_tc. Signed-off-by: Denis Bolotin Signed-off-by: Michal Kalderon Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_dcbx.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c index 8e8fa823d611..69966dfc6e3d 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c @@ -191,7 +191,7 @@ qed_dcbx_dp_protocol(struct qed_hwfn *p_hwfn, struct qed_dcbx_results *p_data) static void qed_dcbx_set_params(struct qed_dcbx_results *p_data, struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, - bool enable, u8 prio, u8 tc, + bool app_tlv, bool enable, u8 prio, u8 tc, enum dcbx_protocol_type type, enum qed_pci_personality personality) { @@ -210,7 +210,7 @@ qed_dcbx_set_params(struct qed_dcbx_results *p_data, p_data->arr[type].dont_add_vlan0 = true; /* QM reconf data */ - if (p_hwfn->hw_info.personality == personality) + if (app_tlv && p_hwfn->hw_info.personality == personality) qed_hw_info_set_offload_tc(&p_hwfn->hw_info, tc); /* Configure dcbx vlan priority in doorbell block for roce EDPM */ @@ -225,7 +225,7 @@ qed_dcbx_set_params(struct qed_dcbx_results *p_data, static void qed_dcbx_update_app_info(struct qed_dcbx_results *p_data, struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, - bool enable, u8 prio, u8 tc, + bool app_tlv, bool enable, u8 prio, u8 tc, enum dcbx_protocol_type type) { enum qed_pci_personality personality; @@ -240,7 +240,7 @@ qed_dcbx_update_app_info(struct qed_dcbx_results *p_data, personality = qed_dcbx_app_update[i].personality; - qed_dcbx_set_params(p_data, p_hwfn, p_ptt, enable, + qed_dcbx_set_params(p_data, p_hwfn, p_ptt, app_tlv, enable, prio, tc, type, personality); } } @@ -319,8 +319,8 @@ qed_dcbx_process_tlv(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, enable = true; } - qed_dcbx_update_app_info(p_data, p_hwfn, p_ptt, enable, - priority, tc, type); + qed_dcbx_update_app_info(p_data, p_hwfn, p_ptt, true, + enable, priority, tc, type); } } @@ -341,7 +341,7 @@ qed_dcbx_process_tlv(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, continue; enable = (type == DCBX_PROTOCOL_ETH) ? false : !!dcbx_version; - qed_dcbx_update_app_info(p_data, p_hwfn, p_ptt, enable, + qed_dcbx_update_app_info(p_data, p_hwfn, p_ptt, false, enable, priority, tc, type); } -- cgit v1.2.3 From 291d57f67d2449737d1e370ab5b9a583818eaa0c Mon Sep 17 00:00:00 2001 From: Michal Kalderon Date: Mon, 12 Nov 2018 12:50:22 +0200 Subject: qed: Fix rdma_info structure allocation Certain flows need to access the rdma-info structure, for example dcbx update flows. In some cases there can be a race between the allocation or deallocation of the structure which was done in roce start / roce stop and an asynchrounous dcbx event that tries to access the structure. For this reason, we move the allocation of the rdma_info structure to be similar to the iscsi/fcoe info structures which are allocated during device setup. We add a new field of "active" to the struct to define whether roce has already been started or not, and this is checked instead of whether the pointer to the info structure. Fixes: 51ff17251c9c ("qed: Add support for RoCE hw init") Signed-off-by: Michal Kalderon Signed-off-by: Denis Bolotin Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_dev.c | 15 ++++++--- drivers/net/ethernet/qlogic/qed/qed_rdma.c | 50 +++++++++++++++++------------- drivers/net/ethernet/qlogic/qed/qed_rdma.h | 5 +++ 3 files changed, 45 insertions(+), 25 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index 7ceb2b97538d..cff141077558 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -185,6 +185,10 @@ void qed_resc_free(struct qed_dev *cdev) qed_iscsi_free(p_hwfn); qed_ooo_free(p_hwfn); } + + if (QED_IS_RDMA_PERSONALITY(p_hwfn)) + qed_rdma_info_free(p_hwfn); + qed_iov_free(p_hwfn); qed_l2_free(p_hwfn); qed_dmae_info_free(p_hwfn); @@ -1081,6 +1085,12 @@ int qed_resc_alloc(struct qed_dev *cdev) goto alloc_err; } + if (QED_IS_RDMA_PERSONALITY(p_hwfn)) { + rc = qed_rdma_info_alloc(p_hwfn); + if (rc) + goto alloc_err; + } + /* DMA info initialization */ rc = qed_dmae_info_alloc(p_hwfn); if (rc) @@ -2102,11 +2112,8 @@ int qed_hw_start_fastpath(struct qed_hwfn *p_hwfn) if (!p_ptt) return -EAGAIN; - /* If roce info is allocated it means roce is initialized and should - * be enabled in searcher. - */ if (p_hwfn->p_rdma_info && - p_hwfn->b_rdma_enabled_in_prs) + p_hwfn->p_rdma_info->active && p_hwfn->b_rdma_enabled_in_prs) qed_wr(p_hwfn, p_ptt, p_hwfn->rdma_prs_search_reg, 0x1); /* Re-open incoming traffic */ diff --git a/drivers/net/ethernet/qlogic/qed/qed_rdma.c b/drivers/net/ethernet/qlogic/qed/qed_rdma.c index 62113438c880..7873d6dfd91f 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_rdma.c +++ b/drivers/net/ethernet/qlogic/qed/qed_rdma.c @@ -140,22 +140,34 @@ static u32 qed_rdma_get_sb_id(void *p_hwfn, u32 rel_sb_id) return FEAT_NUM((struct qed_hwfn *)p_hwfn, QED_PF_L2_QUE) + rel_sb_id; } -static int qed_rdma_alloc(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - struct qed_rdma_start_in_params *params) +int qed_rdma_info_alloc(struct qed_hwfn *p_hwfn) { struct qed_rdma_info *p_rdma_info; - u32 num_cons, num_tasks; - int rc = -ENOMEM; - DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "Allocating RDMA\n"); - - /* Allocate a struct with current pf rdma info */ p_rdma_info = kzalloc(sizeof(*p_rdma_info), GFP_KERNEL); if (!p_rdma_info) - return rc; + return -ENOMEM; + + spin_lock_init(&p_rdma_info->lock); p_hwfn->p_rdma_info = p_rdma_info; + return 0; +} + +void qed_rdma_info_free(struct qed_hwfn *p_hwfn) +{ + kfree(p_hwfn->p_rdma_info); + p_hwfn->p_rdma_info = NULL; +} + +static int qed_rdma_alloc(struct qed_hwfn *p_hwfn) +{ + struct qed_rdma_info *p_rdma_info = p_hwfn->p_rdma_info; + u32 num_cons, num_tasks; + int rc = -ENOMEM; + + DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "Allocating RDMA\n"); + if (QED_IS_IWARP_PERSONALITY(p_hwfn)) p_rdma_info->proto = PROTOCOLID_IWARP; else @@ -183,7 +195,7 @@ static int qed_rdma_alloc(struct qed_hwfn *p_hwfn, /* Allocate a struct with device params and fill it */ p_rdma_info->dev = kzalloc(sizeof(*p_rdma_info->dev), GFP_KERNEL); if (!p_rdma_info->dev) - goto free_rdma_info; + return rc; /* Allocate a struct with port params and fill it */ p_rdma_info->port = kzalloc(sizeof(*p_rdma_info->port), GFP_KERNEL); @@ -298,8 +310,6 @@ free_rdma_port: kfree(p_rdma_info->port); free_rdma_dev: kfree(p_rdma_info->dev); -free_rdma_info: - kfree(p_rdma_info); return rc; } @@ -370,8 +380,6 @@ static void qed_rdma_resc_free(struct qed_hwfn *p_hwfn) kfree(p_rdma_info->port); kfree(p_rdma_info->dev); - - kfree(p_rdma_info); } static void qed_rdma_free_tid(void *rdma_cxt, u32 itid) @@ -679,8 +687,6 @@ static int qed_rdma_setup(struct qed_hwfn *p_hwfn, DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "RDMA setup\n"); - spin_lock_init(&p_hwfn->p_rdma_info->lock); - qed_rdma_init_devinfo(p_hwfn, params); qed_rdma_init_port(p_hwfn); qed_rdma_init_events(p_hwfn, params); @@ -727,7 +733,7 @@ static int qed_rdma_stop(void *rdma_cxt) /* Disable RoCE search */ qed_wr(p_hwfn, p_ptt, p_hwfn->rdma_prs_search_reg, 0); p_hwfn->b_rdma_enabled_in_prs = false; - + p_hwfn->p_rdma_info->active = 0; qed_wr(p_hwfn, p_ptt, PRS_REG_ROCE_DEST_QP_MAX_PF, 0); ll2_ethertype_en = qed_rd(p_hwfn, p_ptt, PRS_REG_LIGHT_L2_ETHERTYPE_EN); @@ -1236,7 +1242,8 @@ qed_rdma_create_qp(void *rdma_cxt, u8 max_stats_queues; int rc; - if (!rdma_cxt || !in_params || !out_params || !p_hwfn->p_rdma_info) { + if (!rdma_cxt || !in_params || !out_params || + !p_hwfn->p_rdma_info->active) { DP_ERR(p_hwfn->cdev, "qed roce create qp failed due to NULL entry (rdma_cxt=%p, in=%p, out=%p, roce_info=?\n", rdma_cxt, in_params, out_params); @@ -1802,8 +1809,8 @@ bool qed_rdma_allocated_qps(struct qed_hwfn *p_hwfn) { bool result; - /* if rdma info has not been allocated, naturally there are no qps */ - if (!p_hwfn->p_rdma_info) + /* if rdma wasn't activated yet, naturally there are no qps */ + if (!p_hwfn->p_rdma_info->active) return false; spin_lock_bh(&p_hwfn->p_rdma_info->lock); @@ -1849,7 +1856,7 @@ static int qed_rdma_start(void *rdma_cxt, if (!p_ptt) goto err; - rc = qed_rdma_alloc(p_hwfn, p_ptt, params); + rc = qed_rdma_alloc(p_hwfn); if (rc) goto err1; @@ -1858,6 +1865,7 @@ static int qed_rdma_start(void *rdma_cxt, goto err2; qed_ptt_release(p_hwfn, p_ptt); + p_hwfn->p_rdma_info->active = 1; return rc; diff --git a/drivers/net/ethernet/qlogic/qed/qed_rdma.h b/drivers/net/ethernet/qlogic/qed/qed_rdma.h index 6f722ee8ee94..50d609c0e108 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_rdma.h +++ b/drivers/net/ethernet/qlogic/qed/qed_rdma.h @@ -102,6 +102,7 @@ struct qed_rdma_info { u16 max_queue_zones; enum protocol_type proto; struct qed_iwarp_info iwarp; + u8 active:1; }; struct qed_rdma_qp { @@ -176,10 +177,14 @@ struct qed_rdma_qp { #if IS_ENABLED(CONFIG_QED_RDMA) void qed_rdma_dpm_bar(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt); void qed_rdma_dpm_conf(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt); +int qed_rdma_info_alloc(struct qed_hwfn *p_hwfn); +void qed_rdma_info_free(struct qed_hwfn *p_hwfn); #else static inline void qed_rdma_dpm_conf(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) {} static inline void qed_rdma_dpm_bar(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) {} +static inline int qed_rdma_info_alloc(struct qed_hwfn *p_hwfn) {return -EINVAL} +static inline void qed_rdma_info_free(struct qed_hwfn *p_hwfn) {} #endif int -- cgit v1.2.3 From ed4eac20dcffdad47709422e0cb925981b056668 Mon Sep 17 00:00:00 2001 From: Denis Bolotin Date: Mon, 12 Nov 2018 12:50:23 +0200 Subject: qed: Fix reading wrong value in loop condition The value of "sb_index" is written by the hardware. Reading its value and writing it to "index" must finish before checking the loop condition. Signed-off-by: Denis Bolotin Signed-off-by: Michal Kalderon Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_int.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.c b/drivers/net/ethernet/qlogic/qed/qed_int.c index 0f0aba793352..b22f464ea3fa 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_int.c +++ b/drivers/net/ethernet/qlogic/qed/qed_int.c @@ -992,6 +992,8 @@ static int qed_int_attentions(struct qed_hwfn *p_hwfn) */ do { index = p_sb_attn->sb_index; + /* finish reading index before the loop condition */ + dma_rmb(); attn_bits = le32_to_cpu(p_sb_attn->atten_bits); attn_acks = le32_to_cpu(p_sb_attn->atten_ack); } while (index != p_sb_attn->sb_index); -- cgit v1.2.3 From 6ba990384e924476b5eed1734f3bcca0df6fd77e Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Thu, 15 Nov 2018 03:25:37 -0500 Subject: bnxt_en: Fix RSS context allocation. Recent commit has added the reservation of RSS context. This requires bnxt_hwrm_vnic_qcaps() to be called before allocating any RSS contexts. The bnxt_hwrm_vnic_qcaps() call sets up proper flags that will determine how many RSS contexts to allocate to support NTUPLE. This causes a regression that too many RSS contexts are being reserved and causing resource shortage when enabling many VFs. Fix it by calling bnxt_hwrm_vnic_qcaps() earlier. Fixes: 41e8d7983752 ("bnxt_en: Modify the ring reservation functions for 57500 series chips.") Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index dd85d790f638..4a45a2b809ea 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -10087,6 +10087,7 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) } bnxt_hwrm_func_qcfg(bp); + bnxt_hwrm_vnic_qcaps(bp); bnxt_hwrm_port_led_qcaps(bp); bnxt_ethtool_init(bp); bnxt_dcb_init(bp); @@ -10120,7 +10121,6 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV6; } - bnxt_hwrm_vnic_qcaps(bp); if (bnxt_rfs_supported(bp)) { dev->hw_features |= NETIF_F_NTUPLE; if (bnxt_rfs_capable(bp)) { -- cgit v1.2.3 From d19819297d9284bd990e22116b8b43d0abcbf488 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Thu, 15 Nov 2018 03:25:38 -0500 Subject: bnxt_en: Fix rx_l4_csum_errors counter on 57500 devices. The software counter structure is defined in both the CP ring's structure and the NQ ring's structure on the new devices. The legacy code adds the counter to the CP ring's structure and the counter won't get displayed since the ethtool code is looking at the NQ ring's structure. Since all other counters are contained in the NQ ring's structure, it makes more sense to count rx_l4_csum_errors in the NQ. Fixes: 50e3ab7836b5 ("bnxt_en: Allocate completion ring structures for 57500 series chips.") Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 4a45a2b809ea..585609990eee 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -1675,7 +1675,7 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr, } else { if (rxcmp1->rx_cmp_cfa_code_errors_v2 & RX_CMP_L4_CS_ERR_BITS) { if (dev->features & NETIF_F_RXCSUM) - cpr->rx_l4_csum_errors++; + bnapi->cp_ring.rx_l4_csum_errors++; } } -- cgit v1.2.3 From addd4df6d763556e16d5316e4e8cd441050cc2af Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Thu, 15 Nov 2018 03:25:39 -0500 Subject: bnxt_en: Disable RDMA support on the 57500 chips. There is no RDMA support on 57500 chips yet, so prevent bnxt_re from registering on these chips. There is intermittent failure if bnxt_re is allowed to register and proceed with RDMA operations. Fixes: 1ab968d2f1d6 ("bnxt_en: Add PCI ID for BCM57508 device.") Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c index beee61292d5e..b59b382d34f9 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c @@ -43,6 +43,9 @@ static int bnxt_register_dev(struct bnxt_en_dev *edev, int ulp_id, if (ulp_id == BNXT_ROCE_ULP) { unsigned int max_stat_ctxs; + if (bp->flags & BNXT_FLAG_CHIP_P5) + return -EOPNOTSUPP; + max_stat_ctxs = bnxt_get_max_func_stat_ctxs(bp); if (max_stat_ctxs <= BNXT_MIN_ROCE_STAT_CTXS || bp->num_stat_ctxs == max_stat_ctxs) -- cgit v1.2.3 From ffd77621700ec3adcf859681e24910c38e0931f5 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Thu, 15 Nov 2018 03:25:40 -0500 Subject: bnxt_en: Workaround occasional TX timeout on 57500 A0. Hardware can sometimes not generate NQ MSIX with a single pending CP ring entry. This seems to always happen at the last entry of the CP ring before it wraps. Add logic to check all the CP rings for pending entries without the CP ring consumer index advancing. Calling HWRM_DBG_RING_INFO_GET to read the context of the CP ring will flush out the NQ entry and MSIX. Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 65 +++++++++++++++++++++++++++++++ drivers/net/ethernet/broadcom/bnxt/bnxt.h | 3 ++ 2 files changed, 68 insertions(+) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 585609990eee..5d4147a75cad 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -8714,6 +8714,26 @@ static int bnxt_set_features(struct net_device *dev, netdev_features_t features) return rc; } +static int bnxt_dbg_hwrm_ring_info_get(struct bnxt *bp, u8 ring_type, + u32 ring_id, u32 *prod, u32 *cons) +{ + struct hwrm_dbg_ring_info_get_output *resp = bp->hwrm_cmd_resp_addr; + struct hwrm_dbg_ring_info_get_input req = {0}; + int rc; + + bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_DBG_RING_INFO_GET, -1, -1); + req.ring_type = ring_type; + req.fw_ring_id = cpu_to_le32(ring_id); + mutex_lock(&bp->hwrm_cmd_lock); + rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); + if (!rc) { + *prod = le32_to_cpu(resp->producer_index); + *cons = le32_to_cpu(resp->consumer_index); + } + mutex_unlock(&bp->hwrm_cmd_lock); + return rc; +} + static void bnxt_dump_tx_sw_state(struct bnxt_napi *bnapi) { struct bnxt_tx_ring_info *txr = bnapi->tx_ring; @@ -8821,6 +8841,11 @@ static void bnxt_timer(struct timer_list *t) bnxt_queue_sp_work(bp); } } + + if ((bp->flags & BNXT_FLAG_CHIP_P5) && netif_carrier_ok(dev)) { + set_bit(BNXT_RING_COAL_NOW_SP_EVENT, &bp->sp_event); + bnxt_queue_sp_work(bp); + } bnxt_restart_timer: mod_timer(&bp->timer, jiffies + bp->current_interval); } @@ -8851,6 +8876,43 @@ static void bnxt_reset(struct bnxt *bp, bool silent) bnxt_rtnl_unlock_sp(bp); } +static void bnxt_chk_missed_irq(struct bnxt *bp) +{ + int i; + + if (!(bp->flags & BNXT_FLAG_CHIP_P5)) + return; + + for (i = 0; i < bp->cp_nr_rings; i++) { + struct bnxt_napi *bnapi = bp->bnapi[i]; + struct bnxt_cp_ring_info *cpr; + u32 fw_ring_id; + int j; + + if (!bnapi) + continue; + + cpr = &bnapi->cp_ring; + for (j = 0; j < 2; j++) { + struct bnxt_cp_ring_info *cpr2 = cpr->cp_ring_arr[j]; + u32 val[2]; + + if (!cpr2 || cpr2->has_more_work || + !bnxt_has_work(bp, cpr2)) + continue; + + if (cpr2->cp_raw_cons != cpr2->last_cp_raw_cons) { + cpr2->last_cp_raw_cons = cpr2->cp_raw_cons; + continue; + } + fw_ring_id = cpr2->cp_ring_struct.fw_ring_id; + bnxt_dbg_hwrm_ring_info_get(bp, + DBG_RING_INFO_GET_REQ_RING_TYPE_L2_CMPL, + fw_ring_id, &val[0], &val[1]); + } + } +} + static void bnxt_cfg_ntp_filters(struct bnxt *); static void bnxt_sp_task(struct work_struct *work) @@ -8930,6 +8992,9 @@ static void bnxt_sp_task(struct work_struct *work) if (test_and_clear_bit(BNXT_FLOW_STATS_SP_EVENT, &bp->sp_event)) bnxt_tc_flow_stats_work(bp); + if (test_and_clear_bit(BNXT_RING_COAL_NOW_SP_EVENT, &bp->sp_event)) + bnxt_chk_missed_irq(bp); + /* These functions below will clear BNXT_STATE_IN_SP_TASK. They * must be the last functions to be called before exiting. */ diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index 498b373c992d..00bd17e55e99 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -798,6 +798,8 @@ struct bnxt_cp_ring_info { u8 had_work_done:1; u8 has_more_work:1; + u32 last_cp_raw_cons; + struct bnxt_coal rx_ring_coal; u64 rx_packets; u64 rx_bytes; @@ -1527,6 +1529,7 @@ struct bnxt { #define BNXT_LINK_SPEED_CHNG_SP_EVENT 14 #define BNXT_FLOW_STATS_SP_EVENT 15 #define BNXT_UPDATE_PHY_SP_EVENT 16 +#define BNXT_RING_COAL_NOW_SP_EVENT 17 struct bnxt_hw_resc hw_resc; struct bnxt_pf_info pf; -- cgit v1.2.3 From 83eb5c5cff32681f3769f502cb5589c7d7509bfe Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Thu, 15 Nov 2018 03:25:41 -0500 Subject: bnxt_en: Add software "missed_irqs" counter. To keep track of the number of times the workaround code for 57500 A0 has been triggered. This is a per NQ counter. Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 1 + drivers/net/ethernet/broadcom/bnxt/bnxt.h | 1 + drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 5 ++++- 3 files changed, 6 insertions(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 5d4147a75cad..d4c300117529 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -8909,6 +8909,7 @@ static void bnxt_chk_missed_irq(struct bnxt *bp) bnxt_dbg_hwrm_ring_info_get(bp, DBG_RING_INFO_GET_REQ_RING_TYPE_L2_CMPL, fw_ring_id, &val[0], &val[1]); + cpr->missed_irqs++; } } } diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index 00bd17e55e99..9e99d4ab3e06 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -818,6 +818,7 @@ struct bnxt_cp_ring_info { dma_addr_t hw_stats_map; u32 hw_stats_ctx_id; u64 rx_l4_csum_errors; + u64 missed_irqs; struct bnxt_ring_struct cp_ring_struct; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index 48078564f025..4b734cd81f8b 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -137,7 +137,7 @@ reset_coalesce: return rc; } -#define BNXT_NUM_STATS 21 +#define BNXT_NUM_STATS 22 #define BNXT_RX_STATS_ENTRY(counter) \ { BNXT_RX_STATS_OFFSET(counter), __stringify(counter) } @@ -384,6 +384,7 @@ static void bnxt_get_ethtool_stats(struct net_device *dev, for (k = 0; k < stat_fields; j++, k++) buf[j] = le64_to_cpu(hw_stats[k]); buf[j++] = cpr->rx_l4_csum_errors; + buf[j++] = cpr->missed_irqs; bnxt_sw_func_stats[RX_TOTAL_DISCARDS].counter += le64_to_cpu(cpr->hw_stats->rx_discard_pkts); @@ -468,6 +469,8 @@ static void bnxt_get_strings(struct net_device *dev, u32 stringset, u8 *buf) buf += ETH_GSTRING_LEN; sprintf(buf, "[%d]: rx_l4_csum_errors", i); buf += ETH_GSTRING_LEN; + sprintf(buf, "[%d]: missed_irqs", i); + buf += ETH_GSTRING_LEN; } for (i = 0; i < BNXT_NUM_SW_FUNC_STATS; i++) { strcpy(buf, bnxt_sw_func_stats[i].string); -- cgit v1.2.3 From 8dc5ae2d48976764cf3498e97963fa06befefb0e Mon Sep 17 00:00:00 2001 From: Vasundhara Volam Date: Thu, 15 Nov 2018 03:25:42 -0500 Subject: bnxt_en: Fix filling time in bnxt_fill_coredump_record() Fix the year and month offset while storing it in bnxt_fill_coredump_record(). Fixes: 6c5657d085ae ("bnxt_en: Add support for ethtool get dump.") Signed-off-by: Vasundhara Volam Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index 4b734cd81f8b..6cc69a58478a 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -2945,8 +2945,8 @@ bnxt_fill_coredump_record(struct bnxt *bp, struct bnxt_coredump_record *record, record->asic_state = 0; strlcpy(record->system_name, utsname()->nodename, sizeof(record->system_name)); - record->year = cpu_to_le16(tm.tm_year); - record->month = cpu_to_le16(tm.tm_mon); + record->year = cpu_to_le16(tm.tm_year + 1900); + record->month = cpu_to_le16(tm.tm_mon + 1); record->day = cpu_to_le16(tm.tm_mday); record->hour = cpu_to_le16(tm.tm_hour); record->minute = cpu_to_le16(tm.tm_min); -- cgit v1.2.3 From ebcd210e93b2a984b7a7b82d45f7f0d21b7ec2d2 Mon Sep 17 00:00:00 2001 From: Ganesh Goudar Date: Thu, 15 Nov 2018 15:36:21 +0530 Subject: cxgb4: fix thermal zone build error with CONFIG_THERMAL=m and cxgb4 as built-in build fails, and 'commit e70a57fa59bb ("cxgb4: fix thermal configuration dependencies")' tries to fix it but when cxgb4i is made built-in build fails again, use IS_REACHABLE instead of IS_ENABLED to fix the issue. Fixes: e70a57fa59bb (cxgb4: fix thermal configuration dependencies) Reported-by: Randy Dunlap Signed-off-by: Ganesh Goudar Signed-off-by: David S. Miller --- drivers/net/ethernet/chelsio/Kconfig | 1 - drivers/net/ethernet/chelsio/cxgb4/Makefile | 4 +--- drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c | 4 ++-- 3 files changed, 3 insertions(+), 6 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/chelsio/Kconfig b/drivers/net/ethernet/chelsio/Kconfig index 75c1c5ed2387..e2cdfa75673f 100644 --- a/drivers/net/ethernet/chelsio/Kconfig +++ b/drivers/net/ethernet/chelsio/Kconfig @@ -67,7 +67,6 @@ config CHELSIO_T3 config CHELSIO_T4 tristate "Chelsio Communications T4/T5/T6 Ethernet support" depends on PCI && (IPV6 || IPV6=n) - depends on THERMAL || !THERMAL select FW_LOADER select MDIO select ZLIB_DEFLATE diff --git a/drivers/net/ethernet/chelsio/cxgb4/Makefile b/drivers/net/ethernet/chelsio/cxgb4/Makefile index 78e5d17a1d5f..91d8a885deba 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/Makefile +++ b/drivers/net/ethernet/chelsio/cxgb4/Makefile @@ -12,6 +12,4 @@ cxgb4-objs := cxgb4_main.o l2t.o smt.o t4_hw.o sge.o clip_tbl.o cxgb4_ethtool.o cxgb4-$(CONFIG_CHELSIO_T4_DCB) += cxgb4_dcb.o cxgb4-$(CONFIG_CHELSIO_T4_FCOE) += cxgb4_fcoe.o cxgb4-$(CONFIG_DEBUG_FS) += cxgb4_debugfs.o -ifdef CONFIG_THERMAL -cxgb4-objs += cxgb4_thermal.o -endif +cxgb4-$(CONFIG_THERMAL) += cxgb4_thermal.o diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index 05a46926016a..d49db46254cd 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -5863,7 +5863,7 @@ fw_attach_fail: if (!is_t4(adapter->params.chip)) cxgb4_ptp_init(adapter); - if (IS_ENABLED(CONFIG_THERMAL) && + if (IS_REACHABLE(CONFIG_THERMAL) && !is_t4(adapter->params.chip) && (adapter->flags & FW_OK)) cxgb4_thermal_init(adapter); @@ -5932,7 +5932,7 @@ static void remove_one(struct pci_dev *pdev) if (!is_t4(adapter->params.chip)) cxgb4_ptp_stop(adapter); - if (IS_ENABLED(CONFIG_THERMAL)) + if (IS_REACHABLE(CONFIG_THERMAL)) cxgb4_thermal_remove(adapter); /* If we allocated filters, free up state associated with any -- cgit v1.2.3 From bd85fbc2038a1bbe84990b23ff69b6fc81a32b2c Mon Sep 17 00:00:00 2001 From: Jack Morgenstein Date: Thu, 15 Nov 2018 18:05:13 +0200 Subject: net/mlx4_core: Zero out lkey field in SW2HW_MPT fw command When re-registering a user mr, the mpt information for the existing mr when running SRIOV is obtained via the QUERY_MPT fw command. The returned information includes the mpt's lkey. This retrieved mpt information is used to move the mpt back to hardware ownership in the rereg flow (via the SW2HW_MPT fw command when running SRIOV). The fw API spec states that for SW2HW_MPT, the lkey field must be zero. Any ConnectX-3 PF driver which checks for strict spec adherence will return failure for SW2HW_MPT if the lkey field is not zero (although the fw in practice ignores this field for SW2HW_MPT). Thus, in order to conform to the fw API spec, set the lkey field to zero before invoking SW2HW_MPT when running SRIOV. Fixes: e630664c8383 ("mlx4_core: Add helper functions to support MR re-registration") Signed-off-by: Jack Morgenstein Signed-off-by: Tariq Toukan Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/mr.c | 1 + 1 file changed, 1 insertion(+) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlx4/mr.c b/drivers/net/ethernet/mellanox/mlx4/mr.c index 2e84f10f59ba..1a11bc0e1612 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mr.c +++ b/drivers/net/ethernet/mellanox/mlx4/mr.c @@ -363,6 +363,7 @@ int mlx4_mr_hw_write_mpt(struct mlx4_dev *dev, struct mlx4_mr *mmr, container_of((void *)mpt_entry, struct mlx4_cmd_mailbox, buf); + (*mpt_entry)->lkey = 0; err = mlx4_SW2HW_MPT(dev, mailbox, key); } -- cgit v1.2.3 From 3ea7e7ea53c9f6ee41cb69a29c375fe9dd9a56a7 Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Thu, 15 Nov 2018 18:05:14 +0200 Subject: net/mlx4_core: Fix uninitialized variable compilation warning Initialize the uid variable to zero to avoid the compilation warning. Fixes: 7a89399ffad7 ("net/mlx4: Add mlx4_bitmap zone allocator") Signed-off-by: Tariq Toukan Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/alloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlx4/alloc.c b/drivers/net/ethernet/mellanox/mlx4/alloc.c index deef5a998985..9af34e03892c 100644 --- a/drivers/net/ethernet/mellanox/mlx4/alloc.c +++ b/drivers/net/ethernet/mellanox/mlx4/alloc.c @@ -337,7 +337,7 @@ void mlx4_zone_allocator_destroy(struct mlx4_zone_allocator *zone_alloc) static u32 __mlx4_alloc_from_zone(struct mlx4_zone_entry *zone, int count, int align, u32 skip_mask, u32 *puid) { - u32 uid; + u32 uid = 0; u32 res; struct mlx4_zone_allocator *zone_alloc = zone->allocator; struct mlx4_zone_entry *curr_node; -- cgit v1.2.3 From a463146e67c848cbab5ce706d6528281b7cded08 Mon Sep 17 00:00:00 2001 From: Aya Levin Date: Thu, 15 Nov 2018 18:05:15 +0200 Subject: net/mlx4: Fix UBSAN warning of signed integer overflow UBSAN: Undefined behavior in drivers/net/ethernet/mellanox/mlx4/resource_tracker.c:626:29 signed integer overflow: 1802201963 + 1802201963 cannot be represented in type 'int' The union of res_reserved and res_port_rsvd[MLX4_MAX_PORTS] monitors granting of reserved resources. The grant operation is calculated and protected, thus both members of the union cannot be negative. Changed type of res_reserved and of res_port_rsvd[MLX4_MAX_PORTS] from signed int to unsigned int, allowing large value. Fixes: 5a0d0a6161ae ("mlx4: Structures and init/teardown for VF resource quotas") Signed-off-by: Aya Levin Signed-off-by: Tariq Toukan Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/mlx4.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h index ebcd2778eeb3..23f1b5b512c2 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h @@ -540,8 +540,8 @@ struct slave_list { struct resource_allocator { spinlock_t alloc_lock; /* protect quotas */ union { - int res_reserved; - int res_port_rsvd[MLX4_MAX_PORTS]; + unsigned int res_reserved; + unsigned int res_port_rsvd[MLX4_MAX_PORTS]; }; union { int res_free; -- cgit v1.2.3 From a97b9565338350d70d8d971c4ee6f0d4fa967418 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 15 Nov 2018 16:15:20 -0800 Subject: drivers/net/ethernet/qlogic/qed/qed_rdma.h: fix typo Add missing semicolon. Fixes: 291d57f67d244973 ("qed: Fix rdma_info structure allocation") Cc: Michal Kalderon Cc: Denis Bolotin Cc: David S. Miller Signed-off-by: Andrew Morton Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_rdma.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/qlogic/qed/qed_rdma.h b/drivers/net/ethernet/qlogic/qed/qed_rdma.h index 50d609c0e108..3689fe3e5935 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_rdma.h +++ b/drivers/net/ethernet/qlogic/qed/qed_rdma.h @@ -183,7 +183,7 @@ void qed_rdma_info_free(struct qed_hwfn *p_hwfn); static inline void qed_rdma_dpm_conf(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) {} static inline void qed_rdma_dpm_bar(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) {} -static inline int qed_rdma_info_alloc(struct qed_hwfn *p_hwfn) {return -EINVAL} +static inline int qed_rdma_info_alloc(struct qed_hwfn *p_hwfn) {return -EINVAL;} static inline void qed_rdma_info_free(struct qed_hwfn *p_hwfn) {} #endif -- cgit v1.2.3 From 83e65df6dfece9eb588735459428f221eb930c0c Mon Sep 17 00:00:00 2001 From: Maxime Chevallier Date: Fri, 9 Nov 2018 09:17:33 +0100 Subject: net: mvneta: Don't advertise 2.5G modes Using 2.5G speed relies on the SerDes lanes being configured accordingly. The lanes have to be reconfigured to switch between 1G and 2.5G, and for now only the bootloader does this configuration. In the case we add a Comphy driver to handle switching the lanes dynamically, it's better for now to stick with supporting only 1G and add advertisement for 2.5G once we really are capable of handling both speeds without problem. Since the interface mode is initialy taken from the DT, we want to make sure that adding comphy support won't break boards that don't update their dtb. Fixes: da58a931f248 ("net: mvneta: Add support for 2500Mbps SGMII") Reported-by: Andrew Lunn Reported-by: Russell King Signed-off-by: Maxime Chevallier Signed-off-by: David S. Miller --- drivers/net/ethernet/marvell/mvneta.c | 12 +++--------- 1 file changed, 3 insertions(+), 9 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c index 3ba672e9e353..e5397c8197b9 100644 --- a/drivers/net/ethernet/marvell/mvneta.c +++ b/drivers/net/ethernet/marvell/mvneta.c @@ -3343,7 +3343,6 @@ static void mvneta_validate(struct net_device *ndev, unsigned long *supported, if (state->interface != PHY_INTERFACE_MODE_NA && state->interface != PHY_INTERFACE_MODE_QSGMII && state->interface != PHY_INTERFACE_MODE_SGMII && - state->interface != PHY_INTERFACE_MODE_2500BASEX && !phy_interface_mode_is_8023z(state->interface) && !phy_interface_mode_is_rgmii(state->interface)) { bitmap_zero(supported, __ETHTOOL_LINK_MODE_MASK_NBITS); @@ -3357,14 +3356,9 @@ static void mvneta_validate(struct net_device *ndev, unsigned long *supported, /* Asymmetric pause is unsupported */ phylink_set(mask, Pause); - /* We cannot use 1Gbps when using the 2.5G interface. */ - if (state->interface == PHY_INTERFACE_MODE_2500BASEX) { - phylink_set(mask, 2500baseT_Full); - phylink_set(mask, 2500baseX_Full); - } else { - phylink_set(mask, 1000baseT_Full); - phylink_set(mask, 1000baseX_Full); - } + /* Half-duplex at speeds higher than 100Mbit is unsupported */ + phylink_set(mask, 1000baseT_Full); + phylink_set(mask, 1000baseX_Full); if (!phy_interface_mode_is_8023z(state->interface)) { /* 10M and 100M are only supported in non-802.3z mode */ -- cgit v1.2.3 From 06bc4d0079ab6a2de86f56703ce1bd13e90b9d9d Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Tue, 13 Nov 2018 18:42:24 +0100 Subject: net: lantiq: Fix returned value in case of error in 'xrx200_probe()' Return 'err' in the error handling path instead of 0. Return explicitly 0 in the normal path, instead of 'err', which is known to be 0 at this point. Fixes: fe1a56420cf2 ("net: lantiq: Add Lantiq / Intel VRX200 Ethernet driver") Signed-off-by: Christophe JAILLET Signed-off-by: David S. Miller --- drivers/net/ethernet/lantiq_xrx200.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/lantiq_xrx200.c b/drivers/net/ethernet/lantiq_xrx200.c index 8c5ba4b81fb7..2d4d10a017e5 100644 --- a/drivers/net/ethernet/lantiq_xrx200.c +++ b/drivers/net/ethernet/lantiq_xrx200.c @@ -512,7 +512,8 @@ static int xrx200_probe(struct platform_device *pdev) err = register_netdev(net_dev); if (err) goto err_unprepare_clk; - return err; + + return 0; err_unprepare_clk: clk_disable_unprepare(priv->clk); @@ -520,7 +521,7 @@ err_unprepare_clk: err_uninit_dma: xrx200_hw_cleanup(priv); - return 0; + return err; } static int xrx200_remove(struct platform_device *pdev) -- cgit v1.2.3 From f2b18732ee9863ac036759baf616ffa03c252ed5 Mon Sep 17 00:00:00 2001 From: Raed Salem Date: Thu, 18 Oct 2018 08:55:21 +0300 Subject: net/mlx5: IPSec, Fix the SA context hash key The commit "net/mlx5: Refactor accel IPSec code" introduced a bug where asynchronous short time change in hash key value by create/release SA context might happen during an asynchronous hash resize operation this could cause a subsequent remove SA context operation to fail as the key value used during resize is not the same key value used when remove SA context operation is invoked. This commit fixes the bug by defining the SA context hash key such that it includes only fields that never change during the lifetime of the SA context object. Fixes: d6c4f0298cec ("net/mlx5: Refactor accel IPSec code") Signed-off-by: Raed Salem Reviewed-by: Aviad Yehezkel Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c index 515e3d6de051..5a22c5874f3b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c @@ -83,8 +83,14 @@ struct mlx5_fpga_ipsec_rule { }; static const struct rhashtable_params rhash_sa = { - .key_len = FIELD_SIZEOF(struct mlx5_fpga_ipsec_sa_ctx, hw_sa), - .key_offset = offsetof(struct mlx5_fpga_ipsec_sa_ctx, hw_sa), + /* Keep out "cmd" field from the key as it's + * value is not constant during the lifetime + * of the key object. + */ + .key_len = FIELD_SIZEOF(struct mlx5_fpga_ipsec_sa_ctx, hw_sa) - + FIELD_SIZEOF(struct mlx5_ifc_fpga_ipsec_sa_v1, cmd), + .key_offset = offsetof(struct mlx5_fpga_ipsec_sa_ctx, hw_sa) + + FIELD_SIZEOF(struct mlx5_ifc_fpga_ipsec_sa_v1, cmd), .head_offset = offsetof(struct mlx5_fpga_ipsec_sa_ctx, hash), .automatic_shrinking = true, .min_size = 1, -- cgit v1.2.3 From acf3766b36d8e59ecbc307894c6d05703ee48014 Mon Sep 17 00:00:00 2001 From: Denis Drozdov Date: Thu, 27 Sep 2018 14:17:54 +0300 Subject: net/mlx5e: IPoIB, Reset QP after channels are closed The mlx5e channels should be closed before mlx5i_uninit_underlay_qp puts the QP into RST (reset) state during mlx5i_close. Currently QP state incorrectly set to RST before channels got deactivated and closed, since mlx5_post_send request expects QP in RTS (Ready To Send) state. The fix is to keep QP in RTS state until mlx5e channels get closed and to reset QP afterwards. Also this fix is simply correct in order to keep the open/close flow symmetric, i.e mlx5i_init_underlay_qp() is called first thing at open, the correct thing to do is to call mlx5i_uninit_underlay_qp() last thing at close, which is exactly what this patch is doing. Fixes: dae37456c8ac ("net/mlx5: Support for attaching multiple underlay QPs to root flow table") Signed-off-by: Denis Drozdov Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c index b59953daf8b4..11dabd62e2c7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c @@ -560,9 +560,9 @@ static int mlx5i_close(struct net_device *netdev) netif_carrier_off(epriv->netdev); mlx5_fs_remove_rx_underlay_qpn(mdev, ipriv->qp.qpn); - mlx5i_uninit_underlay_qp(epriv); mlx5e_deactivate_priv_channels(epriv); mlx5e_close_channels(&epriv->channels); + mlx5i_uninit_underlay_qp(epriv); unlock: mutex_unlock(&epriv->state_lock); return 0; -- cgit v1.2.3 From d3a80bb5a3eac311ddf28387402593977574460d Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Thu, 25 Oct 2018 15:41:58 +0000 Subject: net/mlx5e: Don't match on vlan non-existence if ethertype is wildcarded For the "all" ethertype we should not care whether the packet has vlans. Besides being wrong, the way we did it caused FW error for rules such as: tc filter add dev eth0 protocol all parent ffff: \ prio 1 flower skip_sw action drop b/c the matching meta-data (outer headers bit in struct mlx5_flow_spec) wasn't set. Fix that by matching on vlan non-existence only if we were also told to match on the ethertype. Fixes: cee26487620b ('net/mlx5e: Set vlan masks for all offloaded TC rules') Signed-off-by: Or Gerlitz Reported-by: Slava Ovsiienko Reviewed-by: Jianbo Liu Reviewed-by: Roi Dayan Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 63 +++++++++++++------------ 1 file changed, 32 insertions(+), 31 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 608025ca5c04..84eb6939e69a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -1447,31 +1447,21 @@ static int __parse_cls_flower(struct mlx5e_priv *priv, inner_headers); } - if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_ETH_ADDRS)) { - struct flow_dissector_key_eth_addrs *key = + if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_BASIC)) { + struct flow_dissector_key_basic *key = skb_flow_dissector_target(f->dissector, - FLOW_DISSECTOR_KEY_ETH_ADDRS, + FLOW_DISSECTOR_KEY_BASIC, f->key); - struct flow_dissector_key_eth_addrs *mask = + struct flow_dissector_key_basic *mask = skb_flow_dissector_target(f->dissector, - FLOW_DISSECTOR_KEY_ETH_ADDRS, + FLOW_DISSECTOR_KEY_BASIC, f->mask); + MLX5_SET(fte_match_set_lyr_2_4, headers_c, ethertype, + ntohs(mask->n_proto)); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype, + ntohs(key->n_proto)); - ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, - dmac_47_16), - mask->dst); - ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, - dmac_47_16), - key->dst); - - ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, - smac_47_16), - mask->src); - ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, - smac_47_16), - key->src); - - if (!is_zero_ether_addr(mask->src) || !is_zero_ether_addr(mask->dst)) + if (mask->n_proto) *match_level = MLX5_MATCH_L2; } @@ -1505,9 +1495,10 @@ static int __parse_cls_flower(struct mlx5e_priv *priv, *match_level = MLX5_MATCH_L2; } - } else { + } else if (*match_level != MLX5_MATCH_NONE) { MLX5_SET(fte_match_set_lyr_2_4, headers_c, svlan_tag, 1); MLX5_SET(fte_match_set_lyr_2_4, headers_c, cvlan_tag, 1); + *match_level = MLX5_MATCH_L2; } if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_CVLAN)) { @@ -1545,21 +1536,31 @@ static int __parse_cls_flower(struct mlx5e_priv *priv, } } - if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_BASIC)) { - struct flow_dissector_key_basic *key = + if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_ETH_ADDRS)) { + struct flow_dissector_key_eth_addrs *key = skb_flow_dissector_target(f->dissector, - FLOW_DISSECTOR_KEY_BASIC, + FLOW_DISSECTOR_KEY_ETH_ADDRS, f->key); - struct flow_dissector_key_basic *mask = + struct flow_dissector_key_eth_addrs *mask = skb_flow_dissector_target(f->dissector, - FLOW_DISSECTOR_KEY_BASIC, + FLOW_DISSECTOR_KEY_ETH_ADDRS, f->mask); - MLX5_SET(fte_match_set_lyr_2_4, headers_c, ethertype, - ntohs(mask->n_proto)); - MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype, - ntohs(key->n_proto)); - if (mask->n_proto) + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, + dmac_47_16), + mask->dst); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, + dmac_47_16), + key->dst); + + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, + smac_47_16), + mask->src); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, + smac_47_16), + key->src); + + if (!is_zero_ether_addr(mask->src) || !is_zero_ether_addr(mask->dst)) *match_level = MLX5_MATCH_L2; } -- cgit v1.2.3 From 077ecd785d90c6cbba08d719faa4be8561aa0a1e Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Thu, 18 Oct 2018 12:31:27 +0200 Subject: net/mlx5e: Claim TC hw offloads support only under a proper build config Currently, we are only supporting tc hw offloads when the eswitch support is compiled in, but we are not gating the adevertizment of the NETIF_F_HW_TC feature on this config being set. Fix it, and while doing that, also avoid dealing with the feature on ethtool when the config is not set. Fixes: e8f887ac6a45 ('net/mlx5e: Introduce tc offload support') Signed-off-by: Or Gerlitz Reviewed-by: Roi Dayan Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 1243edbedc9e..fb3b2d9c352b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -3574,6 +3574,7 @@ static int set_feature_cvlan_filter(struct net_device *netdev, bool enable) return 0; } +#ifdef CONFIG_MLX5_ESWITCH static int set_feature_tc_num_filters(struct net_device *netdev, bool enable) { struct mlx5e_priv *priv = netdev_priv(netdev); @@ -3586,6 +3587,7 @@ static int set_feature_tc_num_filters(struct net_device *netdev, bool enable) return 0; } +#endif static int set_feature_rx_all(struct net_device *netdev, bool enable) { @@ -3684,7 +3686,9 @@ static int mlx5e_set_features(struct net_device *netdev, err |= MLX5E_HANDLE_FEATURE(NETIF_F_LRO, set_feature_lro); err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_VLAN_CTAG_FILTER, set_feature_cvlan_filter); +#ifdef CONFIG_MLX5_ESWITCH err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_TC, set_feature_tc_num_filters); +#endif err |= MLX5E_HANDLE_FEATURE(NETIF_F_RXALL, set_feature_rx_all); err |= MLX5E_HANDLE_FEATURE(NETIF_F_RXFCS, set_feature_rx_fcs); err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_VLAN_CTAG_RX, set_feature_rx_vlan); @@ -4678,7 +4682,9 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev) FT_CAP(modify_root) && FT_CAP(identified_miss_table_mode) && FT_CAP(flow_table_modify)) { +#ifdef CONFIG_MLX5_ESWITCH netdev->hw_features |= NETIF_F_HW_TC; +#endif #ifdef CONFIG_MLX5_EN_ARFS netdev->hw_features |= NETIF_F_NTUPLE; #endif -- cgit v1.2.3 From 83621b7df6a646e550fd3d36db2e301cf9a5096b Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Sun, 28 Oct 2018 12:27:29 +0200 Subject: net/mlx5e: Always use the match level enum when parsing TC rule match We get the match level (none, l2, l3, l4) while going over the match dissectors of an offloaded tc rule. When doing this, the match level enum and the not min inline enum values should be used, fix that. This worked accidentally b/c both enums have the same numerical values. Fixes: d708f902989b ('net/mlx5e: Get the required HW match level while parsing TC flow matches') Signed-off-by: Or Gerlitz Reviewed-by: Roi Dayan Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 84eb6939e69a..0409767237a7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -1587,10 +1587,10 @@ static int __parse_cls_flower(struct mlx5e_priv *priv, /* the HW doesn't need L3 inline to match on frag=no */ if (!(key->flags & FLOW_DIS_IS_FRAGMENT)) - *match_level = MLX5_INLINE_MODE_L2; + *match_level = MLX5_MATCH_L2; /* *** L2 attributes parsing up to here *** */ else - *match_level = MLX5_INLINE_MODE_IP; + *match_level = MLX5_MATCH_L3; } } -- cgit v1.2.3 From a1f240f1801721f76bee734c50df2d9529da86e0 Mon Sep 17 00:00:00 2001 From: Yuval Avnery Date: Tue, 16 Oct 2018 15:20:20 -0500 Subject: net/mlx5e: Adjust to max number of channles when re-attaching When core driver enters deattach/attach flow after pci reset, Number of logical CPUs may have changed. As a result we need to update the cpu affiliated resource tables. 1. indirect rqt list 2. eq table Reproduction (PowerPC): echo 1000 > /sys/kernel/debug/powerpc/eeh_max_freezes ppc64_cpu --smt=on # Restart driver modprobe -r ... ; modprobe ... # Link up ifconfig ... # Only physical CPUs ppc64_cpu --smt=off # Inject PCI errors so PCI will reset - calling the pci error handler echo 0x8000000000000000 > /sys/kernel/debug/powerpc//err_injct_inboundA Call trace when trying to add non-existing rqs to an indirect rqt: mlx5e_redirect_rqt+0x84/0x260 [mlx5_core] (unreliable) mlx5e_redirect_rqts+0x188/0x190 [mlx5_core] mlx5e_activate_priv_channels+0x488/0x570 [mlx5_core] mlx5e_open_locked+0xbc/0x140 [mlx5_core] mlx5e_open+0x50/0x130 [mlx5_core] mlx5e_nic_enable+0x174/0x1b0 [mlx5_core] mlx5e_attach_netdev+0x154/0x290 [mlx5_core] mlx5e_attach+0x88/0xd0 [mlx5_core] mlx5_attach_device+0x168/0x1e0 [mlx5_core] mlx5_load_one+0x1140/0x1210 [mlx5_core] mlx5_pci_resume+0x6c/0xf0 [mlx5_core] Create cq will fail when trying to use non-existing EQ. Fixes: 89d44f0a6c73 ("net/mlx5_core: Add pci error handlers to mlx5_core driver") Signed-off-by: Yuval Avnery Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 27 ++++++++++++++++++----- 1 file changed, 22 insertions(+), 5 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index fb3b2d9c352b..25b09bb68e8b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -1623,13 +1623,15 @@ static int mlx5e_alloc_cq_common(struct mlx5_core_dev *mdev, int err; u32 i; + err = mlx5_vector2eqn(mdev, param->eq_ix, &eqn_not_used, &irqn); + if (err) + return err; + err = mlx5_cqwq_create(mdev, ¶m->wq, param->cqc, &cq->wq, &cq->wq_ctrl); if (err) return err; - mlx5_vector2eqn(mdev, param->eq_ix, &eqn_not_used, &irqn); - mcq->cqe_sz = 64; mcq->set_ci_db = cq->wq_ctrl.db.db; mcq->arm_db = cq->wq_ctrl.db.db + 1; @@ -1687,6 +1689,10 @@ static int mlx5e_create_cq(struct mlx5e_cq *cq, struct mlx5e_cq_param *param) int eqn; int err; + err = mlx5_vector2eqn(mdev, param->eq_ix, &eqn, &irqn_not_used); + if (err) + return err; + inlen = MLX5_ST_SZ_BYTES(create_cq_in) + sizeof(u64) * cq->wq_ctrl.buf.npages; in = kvzalloc(inlen, GFP_KERNEL); @@ -1700,8 +1706,6 @@ static int mlx5e_create_cq(struct mlx5e_cq *cq, struct mlx5e_cq_param *param) mlx5_fill_page_frag_array(&cq->wq_ctrl.buf, (__be64 *)MLX5_ADDR_OF(create_cq_in, in, pas)); - mlx5_vector2eqn(mdev, param->eq_ix, &eqn, &irqn_not_used); - MLX5_SET(cqc, cqc, cq_period_mode, param->cq_period_mode); MLX5_SET(cqc, cqc, c_eqn, eqn); MLX5_SET(cqc, cqc, uar_page, mdev->priv.uar->index); @@ -1921,6 +1925,10 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, int err; int eqn; + err = mlx5_vector2eqn(priv->mdev, ix, &eqn, &irq); + if (err) + return err; + c = kvzalloc_node(sizeof(*c), GFP_KERNEL, cpu_to_node(cpu)); if (!c) return -ENOMEM; @@ -1937,7 +1945,6 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, c->xdp = !!params->xdp_prog; c->stats = &priv->channel_stats[ix].ch; - mlx5_vector2eqn(priv->mdev, ix, &eqn, &irq); c->irq_desc = irq_to_desc(irq); netif_napi_add(netdev, &c->napi, mlx5e_napi_poll, 64); @@ -5010,11 +5017,21 @@ err_free_netdev: int mlx5e_attach_netdev(struct mlx5e_priv *priv) { const struct mlx5e_profile *profile; + int max_nch; int err; profile = priv->profile; clear_bit(MLX5E_STATE_DESTROYING, &priv->state); + /* max number of channels may have changed */ + max_nch = mlx5e_get_max_num_channels(priv->mdev); + if (priv->channels.params.num_channels > max_nch) { + mlx5_core_warn(priv->mdev, "MLX5E: Reducing number of channels to %d\n", max_nch); + priv->channels.params.num_channels = max_nch; + mlx5e_build_default_indir_rqt(priv->channels.params.indirection_rqt, + MLX5E_INDIR_RQT_SIZE, max_nch); + } + err = profile->init_tx(priv); if (err) goto out; -- cgit v1.2.3 From 1392f44bba22533b8dfdc6b0e73610b27c3f76cf Mon Sep 17 00:00:00 2001 From: Roi Dayan Date: Tue, 23 Oct 2018 17:30:04 +0300 Subject: net/mlx5e: Apply the correct check for supporting TC esw rules split The mirror and not the output count is the one denoting a split. Fix to condition the offload attempt on the mirror count being > 0 along the firmware to have the related capability. Fixes: 592d36515969 ("net/mlx5e: Parse mirroring action for offloaded TC eswitch flows") Signed-off-by: Roi Dayan Reviewed-by: Yossi Kuperman Reviewed-by: Chris Mi Acked-by: Or Gerlitz Reviewed-by: Or Gerlitz Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 0409767237a7..fca6f4132c91 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -2980,7 +2980,7 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, struct tcf_exts *exts, if (!actions_match_supported(priv, exts, parse_attr, flow, extack)) return -EOPNOTSUPP; - if (attr->out_count > 1 && !mlx5_esw_has_fwd_fdb(priv->mdev)) { + if (attr->mirror_count > 0 && !mlx5_esw_has_fwd_fdb(priv->mdev)) { NL_SET_ERR_MSG_MOD(extack, "current firmware doesn't support split rule for port mirroring"); netdev_warn_once(priv->netdev, "current firmware doesn't support split rule for port mirroring\n"); -- cgit v1.2.3 From 0073c8f72736b423aade8a817587a5f3e4df4ad8 Mon Sep 17 00:00:00 2001 From: Moshe Shemesh Date: Thu, 11 Oct 2018 07:31:10 +0300 Subject: net/mlx5e: RX, verify received packet size in Linear Striding RQ In case of striding RQ, we use MPWRQ (Multi Packet WQE RQ), which means that WQE (RX descriptor) can be used for many packets and so the WQE is much bigger than MTU. In virtualization setups where the port mtu can be larger than the vf mtu, if received packet is bigger than MTU, it won't be dropped by HW on too small receive WQE. If we use linear SKB in striding RQ, since each stride has room for mtu size payload and skb info, an oversized packet can lead to crash for crossing allocated page boundary upon the call to build_skb. So driver needs to check packet size and drop it. Introduce new SW rx counter, rx_oversize_pkts_sw_drop, which counts the number of packets dropped by the driver for being too large. As a new field is added to the RQ struct, re-open the channels whenever this field is being used in datapath (i.e., in the case of linear Striding RQ). Fixes: 619a8f2a42f1 ("net/mlx5e: Use linear SKB in Striding RQ") Signed-off-by: Moshe Shemesh Reviewed-by: Tariq Toukan Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en.h | 1 + drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 4 +++- drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 6 ++++++ drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 3 +++ drivers/net/ethernet/mellanox/mlx5/core/en_stats.h | 2 ++ 5 files changed, 15 insertions(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index d7fbd5b6ac95..118324802926 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -569,6 +569,7 @@ struct mlx5e_rq { unsigned long state; int ix; + unsigned int hw_mtu; struct net_dim dim; /* Dynamic Interrupt Moderation */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 25b09bb68e8b..871313d6b34d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -502,6 +502,7 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c, rq->channel = c; rq->ix = c->ix; rq->mdev = mdev; + rq->hw_mtu = MLX5E_SW2HW_MTU(params, params->sw_mtu); rq->stats = &c->priv->channel_stats[c->ix].rq; rq->xdp_prog = params->xdp_prog ? bpf_prog_inc(params->xdp_prog) : NULL; @@ -3766,10 +3767,11 @@ int mlx5e_change_mtu(struct net_device *netdev, int new_mtu, } if (params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) { + bool is_linear = mlx5e_rx_mpwqe_is_linear_skb(priv->mdev, &new_channels.params); u8 ppw_old = mlx5e_mpwqe_log_pkts_per_wqe(params); u8 ppw_new = mlx5e_mpwqe_log_pkts_per_wqe(&new_channels.params); - reset = reset && (ppw_old != ppw_new); + reset = reset && (is_linear || (ppw_old != ppw_new)); } if (!reset) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index 79638dcbae78..16985ca3248d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -1104,6 +1104,12 @@ mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, u32 frag_size; bool consumed; + /* Check packet size. Note LRO doesn't use linear SKB */ + if (unlikely(cqe_bcnt > rq->hw_mtu)) { + rq->stats->oversize_pkts_sw_drop++; + return NULL; + } + va = page_address(di->page) + head_offset; data = va + rx_headroom; frag_size = MLX5_SKB_FRAG_SZ(rx_headroom + cqe_bcnt32); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c index 1e55b9c27ffc..3e99d0728b2f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c @@ -83,6 +83,7 @@ static const struct counter_desc sw_stats_desc[] = { { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_wqe_err) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_mpwqe_filler_cqes) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_mpwqe_filler_strides) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_oversize_pkts_sw_drop) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_buff_alloc_err) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_cqe_compress_blks) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_cqe_compress_pkts) }, @@ -161,6 +162,7 @@ void mlx5e_grp_sw_update_stats(struct mlx5e_priv *priv) s->rx_wqe_err += rq_stats->wqe_err; s->rx_mpwqe_filler_cqes += rq_stats->mpwqe_filler_cqes; s->rx_mpwqe_filler_strides += rq_stats->mpwqe_filler_strides; + s->rx_oversize_pkts_sw_drop += rq_stats->oversize_pkts_sw_drop; s->rx_buff_alloc_err += rq_stats->buff_alloc_err; s->rx_cqe_compress_blks += rq_stats->cqe_compress_blks; s->rx_cqe_compress_pkts += rq_stats->cqe_compress_pkts; @@ -1189,6 +1191,7 @@ static const struct counter_desc rq_stats_desc[] = { { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, wqe_err) }, { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, mpwqe_filler_cqes) }, { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, mpwqe_filler_strides) }, + { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, oversize_pkts_sw_drop) }, { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, buff_alloc_err) }, { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, cqe_compress_blks) }, { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, cqe_compress_pkts) }, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h index 77f74ce11280..3f8e870ef4c9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h @@ -96,6 +96,7 @@ struct mlx5e_sw_stats { u64 rx_wqe_err; u64 rx_mpwqe_filler_cqes; u64 rx_mpwqe_filler_strides; + u64 rx_oversize_pkts_sw_drop; u64 rx_buff_alloc_err; u64 rx_cqe_compress_blks; u64 rx_cqe_compress_pkts; @@ -193,6 +194,7 @@ struct mlx5e_rq_stats { u64 wqe_err; u64 mpwqe_filler_cqes; u64 mpwqe_filler_strides; + u64 oversize_pkts_sw_drop; u64 buff_alloc_err; u64 cqe_compress_blks; u64 cqe_compress_pkts; -- cgit v1.2.3 From 228c4cd04dfd0667eda182c91504b83c17d97584 Mon Sep 17 00:00:00 2001 From: Valentine Fatiev Date: Wed, 17 Oct 2018 11:45:07 +0300 Subject: net/mlx5e: Fix selftest for small MTUs Loopback test had fixed packet size, which can be bigger than configured MTU. Shorten the loopback packet size to be bigger than minimal MTU allowed by the device. Text field removed from struct 'mlx5ehdr' as redundant to allow send small packets as minimal allowed MTU. Fixes: d605d66 ("net/mlx5e: Add support for ethtool self diagnostics test") Signed-off-by: Valentine Fatiev Reviewed-by: Eran Ben Elisha Signed-off-by: Saeed Mahameed --- .../net/ethernet/mellanox/mlx5/core/en_selftest.c | 26 +++++++++------------- 1 file changed, 10 insertions(+), 16 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c b/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c index 35ded91203f5..4382ef85488c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c @@ -98,18 +98,17 @@ static int mlx5e_test_link_speed(struct mlx5e_priv *priv) return 1; } -#ifdef CONFIG_INET -/* loopback test */ -#define MLX5E_TEST_PKT_SIZE (MLX5E_RX_MAX_HEAD - NET_IP_ALIGN) -static const char mlx5e_test_text[ETH_GSTRING_LEN] = "MLX5E SELF TEST"; -#define MLX5E_TEST_MAGIC 0x5AEED15C001ULL - struct mlx5ehdr { __be32 version; __be64 magic; - char text[ETH_GSTRING_LEN]; }; +#ifdef CONFIG_INET +/* loopback test */ +#define MLX5E_TEST_PKT_SIZE (sizeof(struct ethhdr) + sizeof(struct iphdr) +\ + sizeof(struct udphdr) + sizeof(struct mlx5ehdr)) +#define MLX5E_TEST_MAGIC 0x5AEED15C001ULL + static struct sk_buff *mlx5e_test_get_udp_skb(struct mlx5e_priv *priv) { struct sk_buff *skb = NULL; @@ -117,10 +116,7 @@ static struct sk_buff *mlx5e_test_get_udp_skb(struct mlx5e_priv *priv) struct ethhdr *ethh; struct udphdr *udph; struct iphdr *iph; - int datalen, iplen; - - datalen = MLX5E_TEST_PKT_SIZE - - (sizeof(*ethh) + sizeof(*iph) + sizeof(*udph)); + int iplen; skb = netdev_alloc_skb(priv->netdev, MLX5E_TEST_PKT_SIZE); if (!skb) { @@ -149,7 +145,7 @@ static struct sk_buff *mlx5e_test_get_udp_skb(struct mlx5e_priv *priv) /* Fill UDP header */ udph->source = htons(9); udph->dest = htons(9); /* Discard Protocol */ - udph->len = htons(datalen + sizeof(struct udphdr)); + udph->len = htons(sizeof(struct mlx5ehdr) + sizeof(struct udphdr)); udph->check = 0; /* Fill IP header */ @@ -157,7 +153,8 @@ static struct sk_buff *mlx5e_test_get_udp_skb(struct mlx5e_priv *priv) iph->ttl = 32; iph->version = 4; iph->protocol = IPPROTO_UDP; - iplen = sizeof(struct iphdr) + sizeof(struct udphdr) + datalen; + iplen = sizeof(struct iphdr) + sizeof(struct udphdr) + + sizeof(struct mlx5ehdr); iph->tot_len = htons(iplen); iph->frag_off = 0; iph->saddr = 0; @@ -170,9 +167,6 @@ static struct sk_buff *mlx5e_test_get_udp_skb(struct mlx5e_priv *priv) mlxh = skb_put(skb, sizeof(*mlxh)); mlxh->version = 0; mlxh->magic = cpu_to_be64(MLX5E_TEST_MAGIC); - strlcpy(mlxh->text, mlx5e_test_text, sizeof(mlxh->text)); - datalen -= sizeof(*mlxh); - skb_put_zero(skb, datalen); skb->csum = 0; skb->ip_summed = CHECKSUM_PARTIAL; -- cgit v1.2.3 From e76ad21d070f79e566ac46ce0b0584c3c93e1b43 Mon Sep 17 00:00:00 2001 From: Arthur Kiyanovski Date: Mon, 19 Nov 2018 12:05:20 +0200 Subject: net: ena: fix crash during failed resume from hibernation During resume from hibernation if ena_restore_device fails, ena_com_dev_reset() is called, and uses the readless read mechanism, which was already destroyed by the call to ena_com_mmio_reg_read_request_destroy(). This causes a NULL pointer reference. In this commit we switch the call order of the above two functions to avoid this crash. Fixes: d7703ddbd7c9 ("net: ena: fix rare bug when failed restart/resume is followed by driver removal") Signed-off-by: Arthur Kiyanovski Signed-off-by: David S. Miller --- drivers/net/ethernet/amazon/ena/ena_netdev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c index 18956e7604a3..1d3cead0977e 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.c +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -2694,8 +2694,8 @@ err_device_destroy: ena_com_abort_admin_commands(ena_dev); ena_com_wait_for_abort_completion(ena_dev); ena_com_admin_destroy(ena_dev); - ena_com_mmio_reg_read_request_destroy(ena_dev); ena_com_dev_reset(ena_dev, ENA_REGS_RESET_DRIVER_INVALID_STATE); + ena_com_mmio_reg_read_request_destroy(ena_dev); err: clear_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags); clear_bit(ENA_FLAG_ONGOING_RESET, &adapter->flags); -- cgit v1.2.3 From 58a54b9c62e206b8d5f6e59020bcb178fc271d8e Mon Sep 17 00:00:00 2001 From: Arthur Kiyanovski Date: Mon, 19 Nov 2018 12:05:21 +0200 Subject: net: ena: fix crash during ena_remove() In ena_remove() we have the following stack call: ena_remove() unregister_netdev() ena_destroy_device() netif_carrier_off() Calling netif_carrier_off() causes linkwatch to try to handle the link change event on the already unregistered netdev, which leads to a read from an unreadable memory address. This patch switches the order of the two functions, so that netif_carrier_off() is called on a regiestered netdev. To accomplish this fix we also had to: 1. Remove the set bit ENA_FLAG_TRIGGER_RESET 2. Add a sanitiy check in ena_close() both to prevent double device reset (when calling unregister_netdev() ena_close is called, but the device was already deleted in ena_destroy_device()). 3. Set the admin_queue running state to false to avoid using it after device was reset (for example when calling ena_destroy_all_io_queues() right after ena_com_dev_reset() in ena_down) Fixes: 944b28aa2982 ("net: ena: fix missing lock during device destruction") Signed-off-by: Arthur Kiyanovski Signed-off-by: David S. Miller --- drivers/net/ethernet/amazon/ena/ena_netdev.c | 21 ++++++++++----------- 1 file changed, 10 insertions(+), 11 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c index 1d3cead0977e..a70bb1bb90e7 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.c +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -1848,6 +1848,8 @@ static void ena_down(struct ena_adapter *adapter) rc = ena_com_dev_reset(adapter->ena_dev, adapter->reset_reason); if (rc) dev_err(&adapter->pdev->dev, "Device reset failed\n"); + /* stop submitting admin commands on a device that was reset */ + ena_com_set_admin_running_state(adapter->ena_dev, false); } ena_destroy_all_io_queues(adapter); @@ -1914,6 +1916,9 @@ static int ena_close(struct net_device *netdev) netif_dbg(adapter, ifdown, netdev, "%s\n", __func__); + if (!test_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags)) + return 0; + if (test_bit(ENA_FLAG_DEV_UP, &adapter->flags)) ena_down(adapter); @@ -2613,9 +2618,7 @@ static void ena_destroy_device(struct ena_adapter *adapter, bool graceful) ena_down(adapter); /* Stop the device from sending AENQ events (in case reset flag is set - * and device is up, ena_close already reset the device - * In case the reset flag is set and the device is up, ena_down() - * already perform the reset, so it can be skipped. + * and device is up, ena_down() already reset the device. */ if (!(test_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags) && dev_up)) ena_com_dev_reset(adapter->ena_dev, adapter->reset_reason); @@ -3452,6 +3455,8 @@ err_rss: ena_com_rss_destroy(ena_dev); err_free_msix: ena_com_dev_reset(ena_dev, ENA_REGS_RESET_INIT_ERR); + /* stop submitting admin commands on a device that was reset */ + ena_com_set_admin_running_state(ena_dev, false); ena_free_mgmnt_irq(adapter); ena_disable_msix(adapter); err_worker_destroy: @@ -3498,18 +3503,12 @@ static void ena_remove(struct pci_dev *pdev) cancel_work_sync(&adapter->reset_task); - unregister_netdev(netdev); - - /* If the device is running then we want to make sure the device will be - * reset to make sure no more events will be issued by the device. - */ - if (test_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags)) - set_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags); - rtnl_lock(); ena_destroy_device(adapter, true); rtnl_unlock(); + unregister_netdev(netdev); + free_netdev(netdev); ena_com_rss_destroy(ena_dev); -- cgit v1.2.3 From 4c23738a3f9f203a9b41c89e030eaa8ee241f90f Mon Sep 17 00:00:00 2001 From: Arthur Kiyanovski Date: Mon, 19 Nov 2018 12:05:22 +0200 Subject: net: ena: update driver version from 2.0.1 to 2.0.2 Update driver version due to critical bug fixes. Signed-off-by: Arthur Kiyanovski Signed-off-by: David S. Miller --- drivers/net/ethernet/amazon/ena/ena_netdev.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h index 521873642339..dc8b6173d8d8 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.h +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h @@ -45,7 +45,7 @@ #define DRV_MODULE_VER_MAJOR 2 #define DRV_MODULE_VER_MINOR 0 -#define DRV_MODULE_VER_SUBMINOR 1 +#define DRV_MODULE_VER_SUBMINOR 2 #define DRV_MODULE_NAME "ena" #ifndef DRV_MODULE_VERSION -- cgit v1.2.3 From 9cdeaab3b7e72fe60f582e7658511f30c19f1e29 Mon Sep 17 00:00:00 2001 From: Shay Agroskin Date: Sun, 28 Oct 2018 16:13:46 +0200 Subject: net/mlx5e: Fix a bug in turning off FEC policy in unsupported speeds Some speeds don't support turning FEC policy off. In case a requested FEC policy is not supported for a speed (including current speed), its new FEC policy would be: no FEC - if disabling FEC is supported for that speed unchanged - else Fixes: 2095b2641477 ("net/mlx5e: Add port FEC get/set functions") Signed-off-by: Shay Agroskin Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en/port.c | 28 ++++++++++------------- 1 file changed, 12 insertions(+), 16 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c index 023dc4bccd28..c16351eb9e54 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c @@ -394,12 +394,12 @@ int mlx5e_get_fec_mode(struct mlx5_core_dev *dev, u32 *fec_mode_active, int mlx5e_set_fec_mode(struct mlx5_core_dev *dev, u8 fec_policy) { + u8 fec_policy_nofec = BIT(MLX5E_FEC_NOFEC); bool fec_mode_not_supp_in_speed = false; - u8 no_fec_policy = BIT(MLX5E_FEC_NOFEC); u32 out[MLX5_ST_SZ_DW(pplm_reg)] = {}; u32 in[MLX5_ST_SZ_DW(pplm_reg)] = {}; int sz = MLX5_ST_SZ_BYTES(pplm_reg); - u32 current_fec_speed; + u8 fec_policy_auto = 0; u8 fec_caps = 0; int err; int i; @@ -415,23 +415,19 @@ int mlx5e_set_fec_mode(struct mlx5_core_dev *dev, u8 fec_policy) if (err) return err; - err = mlx5e_port_linkspeed(dev, ¤t_fec_speed); - if (err) - return err; + MLX5_SET(pplm_reg, out, local_port, 1); - memset(in, 0, sz); - MLX5_SET(pplm_reg, in, local_port, 1); - for (i = 0; i < MLX5E_FEC_SUPPORTED_SPEEDS && !!fec_policy; i++) { + for (i = 0; i < MLX5E_FEC_SUPPORTED_SPEEDS; i++) { mlx5e_get_fec_cap_field(out, &fec_caps, fec_supported_speeds[i]); - /* policy supported for link speed */ - if (!!(fec_caps & fec_policy)) { - mlx5e_fec_admin_field(in, &fec_policy, 1, + /* policy supported for link speed, or policy is auto */ + if (fec_caps & fec_policy || fec_policy == fec_policy_auto) { + mlx5e_fec_admin_field(out, &fec_policy, 1, fec_supported_speeds[i]); } else { - if (fec_supported_speeds[i] == current_fec_speed) - return -EOPNOTSUPP; - mlx5e_fec_admin_field(in, &no_fec_policy, 1, - fec_supported_speeds[i]); + /* turn off FEC if supported. Else, leave it the same */ + if (fec_caps & fec_policy_nofec) + mlx5e_fec_admin_field(out, &fec_policy_nofec, 1, + fec_supported_speeds[i]); fec_mode_not_supp_in_speed = true; } } @@ -441,5 +437,5 @@ int mlx5e_set_fec_mode(struct mlx5_core_dev *dev, u8 fec_policy) "FEC policy 0x%x is not supported for some speeds", fec_policy); - return mlx5_core_access_reg(dev, in, sz, out, sz, MLX5_REG_PPLM, 0, 1); + return mlx5_core_access_reg(dev, out, sz, out, sz, MLX5_REG_PPLM, 0, 1); } -- cgit v1.2.3 From febd72f27c02713823cdd011e0087c0e80eb8f62 Mon Sep 17 00:00:00 2001 From: Shay Agroskin Date: Sun, 28 Oct 2018 09:17:29 +0200 Subject: net/mlx5e: Fix wrong field name in FEC related functions This bug would result in reading wrong FEC capabilities for 10G/40G. Fixes: 2095b2641477 ("net/mlx5e: Add port FEC get/set functions") Signed-off-by: Shay Agroskin Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en/port.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c index c16351eb9e54..83ba9ea201d8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c @@ -258,7 +258,7 @@ static int mlx5e_fec_admin_field(u32 *pplm, case 40000: if (!write) *fec_policy = MLX5_GET(pplm_reg, pplm, - fec_override_cap_10g_40g); + fec_override_admin_10g_40g); else MLX5_SET(pplm_reg, pplm, fec_override_admin_10g_40g, *fec_policy); @@ -310,7 +310,7 @@ static int mlx5e_get_fec_cap_field(u32 *pplm, case 10000: case 40000: *fec_cap = MLX5_GET(pplm_reg, pplm, - fec_override_admin_10g_40g); + fec_override_cap_10g_40g); break; case 25000: *fec_cap = MLX5_GET(pplm_reg, pplm, -- cgit v1.2.3 From 64e283348458e2fd2fe41b60dfb6c30e88ee695f Mon Sep 17 00:00:00 2001 From: Shay Agroskin Date: Sun, 28 Oct 2018 09:06:11 +0200 Subject: net/mlx5e: Removed unnecessary warnings in FEC caps query Querying interface FEC caps with 'ethtool [int]' after link reset throws warning regading link speed. This warning is not needed as there is already an indication in user space that the link is not up. Fixes: 0696d60853d5 ("net/mlx5e: Receive buffer configuration") Signed-off-by: Shay Agroskin Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en/port.c | 4 +--- drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c | 4 +++- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c index 83ba9ea201d8..4a37713023be 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c @@ -88,10 +88,8 @@ int mlx5e_port_linkspeed(struct mlx5_core_dev *mdev, u32 *speed) eth_proto_oper = MLX5_GET(ptys_reg, out, eth_proto_oper); *speed = mlx5e_port_ptys2speed(eth_proto_oper); - if (!(*speed)) { - mlx5_core_warn(mdev, "cannot get port speed\n"); + if (!(*speed)) err = -EINVAL; - } return err; } 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 c047da8752da..eac245a93f91 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c @@ -130,8 +130,10 @@ static u32 calculate_xoff(struct mlx5e_priv *priv, unsigned int mtu) int err; err = mlx5e_port_linkspeed(priv->mdev, &speed); - if (err) + if (err) { + mlx5_core_warn(priv->mdev, "cannot get port speed\n"); return 0; + } xoff = (301 + 216 * priv->dcbx.cable_len / 100) * speed / 1000 + 272 * mtu / 100; -- cgit v1.2.3 From 9184e51b5b3ef4509ea869f43e34a60f78f0d32a Mon Sep 17 00:00:00 2001 From: Shay Agroskin Date: Thu, 8 Nov 2018 14:23:36 +0200 Subject: net/mlx5e: Fix failing ethtool query on FEC query error If FEC caps query fails when executing 'ethtool ' the whole callback fails unnecessarily, fixed that by replacing the error return code with debug logging only. Fixes: 6cfa94605091 ("net/mlx5e: Ethtool driver callback for query/set FEC policy") Signed-off-by: Shay Agroskin Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 3e770abfd802..25c1c4f96841 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -843,8 +843,7 @@ static int mlx5e_get_link_ksettings(struct net_device *netdev, ethtool_link_ksettings_add_link_mode(link_ksettings, supported, Autoneg); - err = get_fec_supported_advertised(mdev, link_ksettings); - if (err) + if (get_fec_supported_advertised(mdev, link_ksettings)) netdev_dbg(netdev, "%s: FEC caps query failed: %d\n", __func__, err); -- cgit v1.2.3 From 276d43f0ae963312c0cd0e2b9a85fd11ac65dfcc Mon Sep 17 00:00:00 2001 From: Denis Bolotin Date: Mon, 19 Nov 2018 16:28:30 +0200 Subject: qed: Fix bitmap_weight() check Fix the condition which verifies that only one flag is set. The API bitmap_weight() should receive size in bits instead of bytes. Fixes: b5a9ee7cf3be ("qed: Revise QM cofiguration") Signed-off-by: Denis Bolotin Signed-off-by: Michal Kalderon Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_dev.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index cff141077558..9b41e4ba53e5 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -485,8 +485,11 @@ static u16 *qed_init_qm_get_idx_from_flags(struct qed_hwfn *p_hwfn, struct qed_qm_info *qm_info = &p_hwfn->qm_info; /* Can't have multiple flags set here */ - if (bitmap_weight((unsigned long *)&pq_flags, sizeof(pq_flags)) > 1) + if (bitmap_weight((unsigned long *)&pq_flags, + sizeof(pq_flags) * BITS_PER_BYTE) > 1) { + DP_ERR(p_hwfn, "requested multiple pq flags 0x%x\n", pq_flags); goto err; + } switch (pq_flags) { case PQ_FLAGS_RLS: -- cgit v1.2.3 From eb62cca9bee842e5b23bd0ddfb1f271ca95e8759 Mon Sep 17 00:00:00 2001 From: Denis Bolotin Date: Mon, 19 Nov 2018 16:28:31 +0200 Subject: qed: Fix QM getters to always return a valid pq The getter callers doesn't know the valid Physical Queues (PQ) values. This patch makes sure that a valid PQ will always be returned. The patch consists of 3 fixes: - When qed_init_qm_get_idx_from_flags() receives a disabled flag, it returned PQ 0, which can potentially be another function's pq. Verify that flag is enabled, otherwise return default start_pq. - When qed_init_qm_get_idx_from_flags() receives an unknown flag, it returned NULL and could lead to a segmentation fault. Return default start_pq instead. - A modulo operation was added to MCOS/VFS PQ getters to make sure the PQ returned is in range of the required flag. Fixes: b5a9ee7cf3be ("qed: Revise QM cofiguration") Signed-off-by: Denis Bolotin Signed-off-by: Michal Kalderon Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_dev.c | 24 ++++++++++++++++++++---- 1 file changed, 20 insertions(+), 4 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index 9b41e4ba53e5..88a8576ca9ce 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -491,6 +491,11 @@ static u16 *qed_init_qm_get_idx_from_flags(struct qed_hwfn *p_hwfn, goto err; } + if (!(qed_get_pq_flags(p_hwfn) & pq_flags)) { + DP_ERR(p_hwfn, "pq flag 0x%x is not set\n", pq_flags); + goto err; + } + switch (pq_flags) { case PQ_FLAGS_RLS: return &qm_info->first_rl_pq; @@ -513,8 +518,7 @@ static u16 *qed_init_qm_get_idx_from_flags(struct qed_hwfn *p_hwfn, } err: - DP_ERR(p_hwfn, "BAD pq flags %d\n", pq_flags); - return NULL; + return &qm_info->start_pq; } /* save pq index in qm info */ @@ -538,20 +542,32 @@ u16 qed_get_cm_pq_idx_mcos(struct qed_hwfn *p_hwfn, u8 tc) { u8 max_tc = qed_init_qm_get_num_tcs(p_hwfn); + if (max_tc == 0) { + DP_ERR(p_hwfn, "pq with flag 0x%lx do not exist\n", + PQ_FLAGS_MCOS); + return p_hwfn->qm_info.start_pq; + } + if (tc > max_tc) DP_ERR(p_hwfn, "tc %d must be smaller than %d\n", tc, max_tc); - return qed_get_cm_pq_idx(p_hwfn, PQ_FLAGS_MCOS) + tc; + return qed_get_cm_pq_idx(p_hwfn, PQ_FLAGS_MCOS) + (tc % max_tc); } u16 qed_get_cm_pq_idx_vf(struct qed_hwfn *p_hwfn, u16 vf) { u16 max_vf = qed_init_qm_get_num_vfs(p_hwfn); + if (max_vf == 0) { + DP_ERR(p_hwfn, "pq with flag 0x%lx do not exist\n", + PQ_FLAGS_VFS); + return p_hwfn->qm_info.start_pq; + } + if (vf > max_vf) DP_ERR(p_hwfn, "vf %d must be smaller than %d\n", vf, max_vf); - return qed_get_cm_pq_idx(p_hwfn, PQ_FLAGS_VFS) + vf; + return qed_get_cm_pq_idx(p_hwfn, PQ_FLAGS_VFS) + (vf % max_vf); } u16 qed_get_cm_pq_idx_ofld_mtc(struct qed_hwfn *p_hwfn, u8 tc) -- cgit v1.2.3 From a5681e20b541a507c7d4fd48ae4a4040d32ee1ef Mon Sep 17 00:00:00 2001 From: Juliet Kim Date: Mon, 19 Nov 2018 15:59:22 -0600 Subject: net/ibmnvic: Fix deadlock problem in reset This patch changes to use rtnl_lock only during a reset to avoid deadlock that could occur when a thread operating close is holding rtnl_lock and waiting for reset_lock acquired by another thread, which is waiting for rtnl_lock in order to set the number of tx/rx queues during a reset. Also, we now setting the number of tx/rx queues during a soft reset for failover or LPM events. Signed-off-by: Juliet Kim Signed-off-by: David S. Miller --- drivers/net/ethernet/ibm/ibmvnic.c | 59 ++++++++++++++------------------------ drivers/net/ethernet/ibm/ibmvnic.h | 2 +- 2 files changed, 22 insertions(+), 39 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index c9d5d0a7fbf1..27a6df30eafd 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -1103,20 +1103,15 @@ static int ibmvnic_open(struct net_device *netdev) return 0; } - mutex_lock(&adapter->reset_lock); - if (adapter->state != VNIC_CLOSED) { rc = ibmvnic_login(netdev); - if (rc) { - mutex_unlock(&adapter->reset_lock); + if (rc) return rc; - } rc = init_resources(adapter); if (rc) { netdev_err(netdev, "failed to initialize resources\n"); release_resources(adapter); - mutex_unlock(&adapter->reset_lock); return rc; } } @@ -1124,8 +1119,6 @@ static int ibmvnic_open(struct net_device *netdev) rc = __ibmvnic_open(netdev); netif_carrier_on(netdev); - mutex_unlock(&adapter->reset_lock); - return rc; } @@ -1269,10 +1262,8 @@ static int ibmvnic_close(struct net_device *netdev) return 0; } - mutex_lock(&adapter->reset_lock); rc = __ibmvnic_close(netdev); ibmvnic_cleanup(netdev); - mutex_unlock(&adapter->reset_lock); return rc; } @@ -1820,20 +1811,15 @@ static int do_reset(struct ibmvnic_adapter *adapter, return rc; } else if (adapter->req_rx_queues != old_num_rx_queues || adapter->req_tx_queues != old_num_tx_queues) { - adapter->map_id = 1; release_rx_pools(adapter); release_tx_pools(adapter); - rc = init_rx_pools(netdev); - if (rc) - return rc; - rc = init_tx_pools(netdev); - if (rc) - return rc; - release_napi(adapter); - rc = init_napi(adapter); + release_vpd_data(adapter); + + rc = init_resources(adapter); if (rc) return rc; + } else { rc = reset_tx_pools(adapter); if (rc) @@ -1917,17 +1903,8 @@ static int do_hard_reset(struct ibmvnic_adapter *adapter, adapter->state = VNIC_PROBED; return 0; } - /* netif_set_real_num_xx_queues needs to take rtnl lock here - * unless wait_for_reset is set, in which case the rtnl lock - * has already been taken before initializing the reset - */ - if (!adapter->wait_for_reset) { - rtnl_lock(); - rc = init_resources(adapter); - rtnl_unlock(); - } else { - rc = init_resources(adapter); - } + + rc = init_resources(adapter); if (rc) return rc; @@ -1986,13 +1963,21 @@ static void __ibmvnic_reset(struct work_struct *work) struct ibmvnic_rwi *rwi; struct ibmvnic_adapter *adapter; struct net_device *netdev; + bool we_lock_rtnl = false; u32 reset_state; int rc = 0; adapter = container_of(work, struct ibmvnic_adapter, ibmvnic_reset); netdev = adapter->netdev; - mutex_lock(&adapter->reset_lock); + /* netif_set_real_num_xx_queues needs to take rtnl lock here + * unless wait_for_reset is set, in which case the rtnl lock + * has already been taken before initializing the reset + */ + if (!adapter->wait_for_reset) { + rtnl_lock(); + we_lock_rtnl = true; + } reset_state = adapter->state; rwi = get_next_rwi(adapter); @@ -2020,12 +2005,11 @@ static void __ibmvnic_reset(struct work_struct *work) if (rc) { netdev_dbg(adapter->netdev, "Reset failed\n"); free_all_rwi(adapter); - mutex_unlock(&adapter->reset_lock); - return; } adapter->resetting = false; - mutex_unlock(&adapter->reset_lock); + if (we_lock_rtnl) + rtnl_unlock(); } static int ibmvnic_reset(struct ibmvnic_adapter *adapter, @@ -4768,7 +4752,6 @@ static int ibmvnic_probe(struct vio_dev *dev, const struct vio_device_id *id) INIT_WORK(&adapter->ibmvnic_reset, __ibmvnic_reset); INIT_LIST_HEAD(&adapter->rwi_list); - mutex_init(&adapter->reset_lock); mutex_init(&adapter->rwi_lock); adapter->resetting = false; @@ -4840,8 +4823,8 @@ static int ibmvnic_remove(struct vio_dev *dev) struct ibmvnic_adapter *adapter = netdev_priv(netdev); adapter->state = VNIC_REMOVING; - unregister_netdev(netdev); - mutex_lock(&adapter->reset_lock); + rtnl_lock(); + unregister_netdevice(netdev); release_resources(adapter); release_sub_crqs(adapter, 1); @@ -4852,7 +4835,7 @@ static int ibmvnic_remove(struct vio_dev *dev) adapter->state = VNIC_REMOVED; - mutex_unlock(&adapter->reset_lock); + rtnl_unlock(); device_remove_file(&dev->dev, &dev_attr_failover); free_netdev(netdev); dev_set_drvdata(&dev->dev, NULL); diff --git a/drivers/net/ethernet/ibm/ibmvnic.h b/drivers/net/ethernet/ibm/ibmvnic.h index 18103b811d4d..99c4f8d331ce 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.h +++ b/drivers/net/ethernet/ibm/ibmvnic.h @@ -1075,7 +1075,7 @@ struct ibmvnic_adapter { struct tasklet_struct tasklet; enum vnic_state state; enum ibmvnic_reset_reason reset_reason; - struct mutex reset_lock, rwi_lock; + struct mutex rwi_lock; struct list_head rwi_list; struct work_struct ibmvnic_reset; bool resetting; -- cgit v1.2.3 From 59663e42199c93d1d7314d1446f6782fc4b1eb81 Mon Sep 17 00:00:00 2001 From: Siva Reddy Kallam Date: Tue, 20 Nov 2018 10:04:04 +0530 Subject: tg3: Add PHY reset for 5717/5719/5720 in change ring and flow control paths This patch has the fix to avoid PHY lockup with 5717/5719/5720 in change ring and flow control paths. This patch solves the RX hang while doing continuous ring or flow control parameters with heavy traffic from peer. Signed-off-by: Siva Reddy Kallam Acked-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/tg3.c | 18 ++++++++++++++++-- 1 file changed, 16 insertions(+), 2 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index 89295306f161..432c3b867084 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -12422,6 +12422,7 @@ static int tg3_set_ringparam(struct net_device *dev, struct ethtool_ringparam *e { struct tg3 *tp = netdev_priv(dev); int i, irq_sync = 0, err = 0; + bool reset_phy = false; if ((ering->rx_pending > tp->rx_std_ring_mask) || (ering->rx_jumbo_pending > tp->rx_jmb_ring_mask) || @@ -12453,7 +12454,13 @@ static int tg3_set_ringparam(struct net_device *dev, struct ethtool_ringparam *e if (netif_running(dev)) { tg3_halt(tp, RESET_KIND_SHUTDOWN, 1); - err = tg3_restart_hw(tp, false); + /* Reset PHY to avoid PHY lock up */ + if (tg3_asic_rev(tp) == ASIC_REV_5717 || + tg3_asic_rev(tp) == ASIC_REV_5719 || + tg3_asic_rev(tp) == ASIC_REV_5720) + reset_phy = true; + + err = tg3_restart_hw(tp, reset_phy); if (!err) tg3_netif_start(tp); } @@ -12487,6 +12494,7 @@ static int tg3_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam { struct tg3 *tp = netdev_priv(dev); int err = 0; + bool reset_phy = false; if (tp->link_config.autoneg == AUTONEG_ENABLE) tg3_warn_mgmt_link_flap(tp); @@ -12556,7 +12564,13 @@ static int tg3_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam if (netif_running(dev)) { tg3_halt(tp, RESET_KIND_SHUTDOWN, 1); - err = tg3_restart_hw(tp, false); + /* Reset PHY to avoid PHY lock up */ + if (tg3_asic_rev(tp) == ASIC_REV_5717 || + tg3_asic_rev(tp) == ASIC_REV_5719 || + tg3_asic_rev(tp) == ASIC_REV_5720) + reset_phy = true; + + err = tg3_restart_hw(tp, reset_phy); if (!err) tg3_netif_start(tp); } -- cgit v1.2.3 From 426a593e641ebf0d9288f0a2fcab644a86820220 Mon Sep 17 00:00:00 2001 From: Vincent Chen Date: Wed, 21 Nov 2018 09:38:11 +0800 Subject: net: faraday: ftmac100: remove netif_running(netdev) check before disabling interrupts In the original ftmac100_interrupt(), the interrupts are only disabled when the condition "netif_running(netdev)" is true. However, this condition causes kerenl hang in the following case. When the user requests to disable the network device, kernel will clear the bit __LINK_STATE_START from the dev->state and then call the driver's ndo_stop function. Network device interrupts are not blocked during this process. If an interrupt occurs between clearing __LINK_STATE_START and stopping network device, kernel cannot disable the interrupts due to the condition "netif_running(netdev)" in the ISR. Hence, kernel will hang due to the continuous interruption of the network device. In order to solve the above problem, the interrupts of the network device should always be disabled in the ISR without being restricted by the condition "netif_running(netdev)". [V2] Remove unnecessary curly braces. Signed-off-by: Vincent Chen Signed-off-by: David S. Miller --- drivers/net/ethernet/faraday/ftmac100.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/faraday/ftmac100.c b/drivers/net/ethernet/faraday/ftmac100.c index 570caeb8ee9e..084f24daf2b5 100644 --- a/drivers/net/ethernet/faraday/ftmac100.c +++ b/drivers/net/ethernet/faraday/ftmac100.c @@ -872,11 +872,10 @@ static irqreturn_t ftmac100_interrupt(int irq, void *dev_id) struct net_device *netdev = dev_id; struct ftmac100 *priv = netdev_priv(netdev); - if (likely(netif_running(netdev))) { - /* Disable interrupts for polling */ - ftmac100_disable_all_int(priv); + /* Disable interrupts for polling */ + ftmac100_disable_all_int(priv); + if (likely(netif_running(netdev))) napi_schedule(&priv->napi); - } return IRQ_HANDLED; } -- cgit v1.2.3 From 6d0f60b0f8588fd4380ea5df9601e12fddd55ce2 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Wed, 21 Nov 2018 16:32:10 +0100 Subject: net: thunderx: set xdp_prog to NULL if bpf_prog_add fails Set xdp_prog pointer to NULL if bpf_prog_add fails since that routine reports the error code instead of NULL in case of failure and xdp_prog pointer value is used in the driver to verify if XDP is currently enabled. Moreover report the error code to userspace if nicvf_xdp_setup fails Fixes: 05c773f52b96 ("net: thunderx: Add basic XDP support") Signed-off-by: Lorenzo Bianconi Signed-off-by: David S. Miller --- drivers/net/ethernet/cavium/thunder/nicvf_main.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c index 768f584f8392..88f8a8fa93cd 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c @@ -1784,6 +1784,7 @@ static int nicvf_xdp_setup(struct nicvf *nic, struct bpf_prog *prog) bool if_up = netif_running(nic->netdev); struct bpf_prog *old_prog; bool bpf_attached = false; + int ret = 0; /* For now just support only the usual MTU sized frames */ if (prog && (dev->mtu > 1500)) { @@ -1817,8 +1818,12 @@ static int nicvf_xdp_setup(struct nicvf *nic, struct bpf_prog *prog) if (nic->xdp_prog) { /* Attach BPF program */ nic->xdp_prog = bpf_prog_add(nic->xdp_prog, nic->rx_queues - 1); - if (!IS_ERR(nic->xdp_prog)) + if (!IS_ERR(nic->xdp_prog)) { bpf_attached = true; + } else { + ret = PTR_ERR(nic->xdp_prog); + nic->xdp_prog = NULL; + } } /* Calculate Tx queues needed for XDP and network stack */ @@ -1830,7 +1835,7 @@ static int nicvf_xdp_setup(struct nicvf *nic, struct bpf_prog *prog) netif_trans_update(nic->netdev); } - return 0; + return ret; } static int nicvf_xdp(struct net_device *netdev, struct netdev_bpf *xdp) -- cgit v1.2.3 From b7cdec3d699db2e5985ad39de0f25d3b6111928e Mon Sep 17 00:00:00 2001 From: Thomas Falcon Date: Wed, 21 Nov 2018 11:17:58 -0600 Subject: ibmvnic: Fix RX queue buffer cleanup The wrong index is used when cleaning up RX buffer objects during release of RX queues. Update to use the correct index counter. Signed-off-by: Thomas Falcon Signed-off-by: David S. Miller --- drivers/net/ethernet/ibm/ibmvnic.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index 27a6df30eafd..066897a350a5 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -485,8 +485,8 @@ static void release_rx_pools(struct ibmvnic_adapter *adapter) for (j = 0; j < rx_pool->size; j++) { if (rx_pool->rx_buff[j].skb) { - dev_kfree_skb_any(rx_pool->rx_buff[i].skb); - rx_pool->rx_buff[i].skb = NULL; + dev_kfree_skb_any(rx_pool->rx_buff[j].skb); + rx_pool->rx_buff[j].skb = NULL; } } -- cgit v1.2.3 From 5bf032ef08e6a110edc1e3bfb3c66a208fb55125 Mon Sep 17 00:00:00 2001 From: Thomas Falcon Date: Wed, 21 Nov 2018 11:17:59 -0600 Subject: ibmvnic: Update driver queues after change in ring size support During device reset, queue memory is not being updated to accommodate changes in ring buffer sizes supported by backing hardware. Track any differences in ring buffer sizes following the reset and update queue memory when possible. Signed-off-by: Thomas Falcon Signed-off-by: David S. Miller --- drivers/net/ethernet/ibm/ibmvnic.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index 066897a350a5..c0203a0d5e3b 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -1737,6 +1737,7 @@ static int do_reset(struct ibmvnic_adapter *adapter, struct ibmvnic_rwi *rwi, u32 reset_state) { u64 old_num_rx_queues, old_num_tx_queues; + u64 old_num_rx_slots, old_num_tx_slots; struct net_device *netdev = adapter->netdev; int i, rc; @@ -1748,6 +1749,8 @@ static int do_reset(struct ibmvnic_adapter *adapter, old_num_rx_queues = adapter->req_rx_queues; old_num_tx_queues = adapter->req_tx_queues; + old_num_rx_slots = adapter->req_rx_add_entries_per_subcrq; + old_num_tx_slots = adapter->req_tx_entries_per_subcrq; ibmvnic_cleanup(netdev); @@ -1810,7 +1813,11 @@ static int do_reset(struct ibmvnic_adapter *adapter, if (rc) return rc; } else if (adapter->req_rx_queues != old_num_rx_queues || - adapter->req_tx_queues != old_num_tx_queues) { + adapter->req_tx_queues != old_num_tx_queues || + adapter->req_rx_add_entries_per_subcrq != + old_num_rx_slots || + adapter->req_tx_entries_per_subcrq != + old_num_tx_slots) { release_rx_pools(adapter); release_tx_pools(adapter); release_napi(adapter); -- cgit v1.2.3 From c44c749d3b6fdfca39002e7e48e03fe9f9fe37a3 Mon Sep 17 00:00:00 2001 From: Yangtao Li Date: Thu, 22 Nov 2018 07:34:41 -0500 Subject: net: amd: add missing of_node_put() of_find_node_by_path() acquires a reference to the node returned by it and that reference needs to be dropped by its caller. This place doesn't do that, so fix it. Signed-off-by: Yangtao Li Signed-off-by: David S. Miller --- drivers/net/ethernet/amd/sunlance.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/amd/sunlance.c b/drivers/net/ethernet/amd/sunlance.c index b4fc0ed5bce8..9d4899826823 100644 --- a/drivers/net/ethernet/amd/sunlance.c +++ b/drivers/net/ethernet/amd/sunlance.c @@ -1419,7 +1419,7 @@ static int sparc_lance_probe_one(struct platform_device *op, prop = of_get_property(nd, "tpe-link-test?", NULL); if (!prop) - goto no_link_test; + goto node_put; if (strcmp(prop, "true")) { printk(KERN_NOTICE "SunLance: warning: overriding option " @@ -1428,6 +1428,8 @@ static int sparc_lance_probe_one(struct platform_device *op, "to ecd@skynet.be\n"); auxio_set_lte(AUXIO_LTE_ON); } +node_put: + of_node_put(nd); no_link_test: lp->auto_select = 1; lp->tpe = 0; -- cgit v1.2.3 From ef2a7cf1d8831535b8991459567b385661eb4a36 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 23 Nov 2018 18:28:01 +0100 Subject: net: thunderx: set tso_hdrs pointer to NULL in nicvf_free_snd_queue Reset snd_queue tso_hdrs pointer to NULL in nicvf_free_snd_queue routine since it is used to check if tso dma descriptor queue has been previously allocated. The issue can be triggered with the following reproducer: $ip link set dev enP2p1s0v0 xdpdrv obj xdp_dummy.o $ip link set dev enP2p1s0v0 xdpdrv off [ 341.467649] WARNING: CPU: 74 PID: 2158 at mm/vmalloc.c:1511 __vunmap+0x98/0xe0 [ 341.515010] Hardware name: GIGABYTE H270-T70/MT70-HD0, BIOS T49 02/02/2018 [ 341.521874] pstate: 60400005 (nZCv daif +PAN -UAO) [ 341.526654] pc : __vunmap+0x98/0xe0 [ 341.530132] lr : __vunmap+0x98/0xe0 [ 341.533609] sp : ffff00001c5db860 [ 341.536913] x29: ffff00001c5db860 x28: 0000000000020000 [ 341.542214] x27: ffff810feb5090b0 x26: ffff000017e57000 [ 341.547515] x25: 0000000000000000 x24: 00000000fbd00000 [ 341.552816] x23: 0000000000000000 x22: ffff810feb5090b0 [ 341.558117] x21: 0000000000000000 x20: 0000000000000000 [ 341.563418] x19: ffff000017e57000 x18: 0000000000000000 [ 341.568719] x17: 0000000000000000 x16: 0000000000000000 [ 341.574020] x15: 0000000000000010 x14: ffffffffffffffff [ 341.579321] x13: ffff00008985eb27 x12: ffff00000985eb2f [ 341.584622] x11: ffff0000096b3000 x10: ffff00001c5db510 [ 341.589923] x9 : 00000000ffffffd0 x8 : ffff0000086868e8 [ 341.595224] x7 : 3430303030303030 x6 : 00000000000006ef [ 341.600525] x5 : 00000000003fffff x4 : 0000000000000000 [ 341.605825] x3 : 0000000000000000 x2 : ffffffffffffffff [ 341.611126] x1 : ffff0000096b3728 x0 : 0000000000000038 [ 341.616428] Call trace: [ 341.618866] __vunmap+0x98/0xe0 [ 341.621997] vunmap+0x3c/0x50 [ 341.624961] arch_dma_free+0x68/0xa0 [ 341.628534] dma_direct_free+0x50/0x80 [ 341.632285] nicvf_free_resources+0x160/0x2d8 [nicvf] [ 341.637327] nicvf_config_data_transfer+0x174/0x5e8 [nicvf] [ 341.642890] nicvf_stop+0x298/0x340 [nicvf] [ 341.647066] __dev_close_many+0x9c/0x108 [ 341.650977] dev_close_many+0xa4/0x158 [ 341.654720] rollback_registered_many+0x140/0x530 [ 341.659414] rollback_registered+0x54/0x80 [ 341.663499] unregister_netdevice_queue+0x9c/0xe8 [ 341.668192] unregister_netdev+0x28/0x38 [ 341.672106] nicvf_remove+0xa4/0xa8 [nicvf] [ 341.676280] nicvf_shutdown+0x20/0x30 [nicvf] [ 341.680630] pci_device_shutdown+0x44/0x88 [ 341.684720] device_shutdown+0x144/0x250 [ 341.688640] kernel_restart_prepare+0x44/0x50 [ 341.692986] kernel_restart+0x20/0x68 [ 341.696638] __se_sys_reboot+0x210/0x238 [ 341.700550] __arm64_sys_reboot+0x24/0x30 [ 341.704555] el0_svc_handler+0x94/0x110 [ 341.708382] el0_svc+0x8/0xc [ 341.711252] ---[ end trace 3f4019c8439959c9 ]--- [ 341.715874] page:ffff7e0003ef4000 count:0 mapcount:0 mapping:0000000000000000 index:0x4 [ 341.723872] flags: 0x1fffe000000000() [ 341.727527] raw: 001fffe000000000 ffff7e0003f1a008 ffff7e0003ef4048 0000000000000000 [ 341.735263] raw: 0000000000000004 0000000000000000 00000000ffffffff 0000000000000000 [ 341.742994] page dumped because: VM_BUG_ON_PAGE(page_ref_count(page) == 0) where xdp_dummy.c is a simple bpf program that forwards the incoming frames to the network stack (available here: https://github.com/altoor/xdp_walkthrough_examples/blob/master/sample_1/xdp_dummy.c) Fixes: 05c773f52b96 ("net: thunderx: Add basic XDP support") Fixes: 4863dea3fab0 ("net: Adding support for Cavium ThunderX network controller") Signed-off-by: Lorenzo Bianconi Signed-off-by: David S. Miller --- drivers/net/ethernet/cavium/thunder/nicvf_queues.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c index 187a249ff2d1..fcaf18fa3904 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c @@ -585,10 +585,12 @@ static void nicvf_free_snd_queue(struct nicvf *nic, struct snd_queue *sq) if (!sq->dmem.base) return; - if (sq->tso_hdrs) + if (sq->tso_hdrs) { dma_free_coherent(&nic->pdev->dev, sq->dmem.q_len * TSO_HEADER_SIZE, sq->tso_hdrs, sq->tso_hdrs_phys); + sq->tso_hdrs = NULL; + } /* Free pending skbs in the queue */ smp_rmb(); -- cgit v1.2.3 From 07093b76476903f820d83d56c3040e656fb4d9e3 Mon Sep 17 00:00:00 2001 From: Andreas Fiedler Date: Sat, 24 Nov 2018 00:16:34 +0100 Subject: net: gemini: Fix copy/paste error The TX stats should be started with the tx_stats_syncp, there seems to be a copy/paste error in the driver. Signed-off-by: Andreas Fiedler Signed-off-by: Linus Walleij Signed-off-by: David S. Miller --- drivers/net/ethernet/cortina/gemini.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet') diff --git a/drivers/net/ethernet/cortina/gemini.c b/drivers/net/ethernet/cortina/gemini.c index ceec467f590d..949103db8a8a 100644 --- a/drivers/net/ethernet/cortina/gemini.c +++ b/drivers/net/ethernet/cortina/gemini.c @@ -660,7 +660,7 @@ static void gmac_clean_txq(struct net_device *netdev, struct gmac_txq *txq, u64_stats_update_begin(&port->tx_stats_syncp); port->tx_frag_stats[nfrags]++; - u64_stats_update_end(&port->ir_stats_syncp); + u64_stats_update_end(&port->tx_stats_syncp); } } -- cgit v1.2.3