From d1ec975f9fa6d2211c1f403010361034a87e317f Mon Sep 17 00:00:00 2001 From: Maciej Fijalkowski Date: Mon, 29 Nov 2021 15:17:46 -0800 Subject: ice: xsk: clear status_error0 for each allocated desc Fix a bug in which the receiving of packets can stop in the zero-copy driver. Ice HW ignores 3 lower bits from QRX_TAIL register, which means that tail is bumped only on intervals of 8. Currently with XSK RX batching in place, ice_alloc_rx_bufs_zc() clears the status_error0 only of the last descriptor that has been allocated/taken from the XSK buffer pool. status_error0 includes DD bit that is looked upon by the ice_clean_rx_irq_zc() to tell if a descriptor can be processed. The bug can be triggered when driver updates the ntu but not the QRX_TAIL, so HW wouldn't have a chance to write to the ready descriptors. Later on driver moves the ntc to the mentioned set of descriptors and interprets them as a ready to be processed, since corresponding DD bits were not cleared nor any writeback has happened that would clear it. This can then lead to ntc == ntu case which means that ring is empty and no further packet processing. Fix the XSK traffic hang that can be observed when l2fwd scenario from xdpsock is used by making sure that status_error0 is cleared for each descriptor that is fed to HW and therefore we are sure that driver will not processed non-valid DD bits. This will also prevent the driver from processing the descriptors that were allocated in favor of the previously processed ones, but writeback didn't happen yet. Fixes: db804cfc21e9 ("ice: Use the xsk batched rx allocation interface") Signed-off-by: Maciej Fijalkowski Reviewed-by: Alexander Lobakin Signed-off-by: Tony Nguyen Signed-off-by: David S. Miller --- drivers/net/ethernet/intel/ice/ice_xsk.c | 1 + 1 file changed, 1 insertion(+) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c index ff55cb415b11..bb9a80847298 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.c +++ b/drivers/net/ethernet/intel/ice/ice_xsk.c @@ -383,6 +383,7 @@ bool ice_alloc_rx_bufs_zc(struct ice_rx_ring *rx_ring, u16 count) while (i--) { dma = xsk_buff_xdp_get_dma(*xdp); rx_desc->read.pkt_addr = cpu_to_le64(dma); + rx_desc->wb.status_error0 = 0; rx_desc++; xdp++; -- cgit v1.2.3 From 7e4dcc13965c57869684d57a1dc6dd7be589488c Mon Sep 17 00:00:00 2001 From: Mitch Williams Date: Fri, 4 Jun 2021 09:53:28 -0700 Subject: iavf: restore MSI state on reset If the PF experiences an FLR, the VF's MSI and MSI-X configuration will be conveniently and silently removed in the process. When this happens, reset recovery will appear to complete normally but no traffic will pass. The netdev watchdog will helpfully notify everyone of this issue. To prevent such public embarrassment, restore MSI configuration at every reset. For normal resets, this will do no harm, but for VF resets resulting from a PF FLR, this will keep the VF working. Fixes: 5eae00c57f5e ("i40evf: main driver core") Signed-off-by: Mitch Williams Tested-by: George Kuruvinakunnel Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/iavf/iavf_main.c | 1 + 1 file changed, 1 insertion(+) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c index 14934a7a13ef..cfdbf8c08d18 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_main.c +++ b/drivers/net/ethernet/intel/iavf/iavf_main.c @@ -2248,6 +2248,7 @@ static void iavf_reset_task(struct work_struct *work) } pci_set_master(adapter->pdev); + pci_restore_msi_state(adapter->pdev); if (i == IAVF_RESET_WAIT_COMPLETE_COUNT) { dev_err(&adapter->pdev->dev, "Reset never finished (%x)\n", -- cgit v1.2.3 From 1a1aa356ddf3f16539f5962c01c5f702686dfc15 Mon Sep 17 00:00:00 2001 From: Michal Maloszewski Date: Tue, 26 Oct 2021 12:59:09 +0000 Subject: iavf: Fix reporting when setting descriptor count iavf_set_ringparams doesn't communicate to the user that 1. The user requested descriptor count is out of range. Instead it just quietly sets descriptors to the "clamped" value and calls it done. This makes it look an invalid value was successfully set as the descriptor count when this isn't actually true. 2. The user provided descriptor count needs to be inflated for alignment reasons. This behavior is confusing. The ice driver has already addressed this by rejecting invalid values for descriptor count and messaging for alignment adjustments. Do the same thing here by adding the error and info messages. Fixes: fbb7ddfef253 ("i40evf: core ethtool functionality") Signed-off-by: Anirudh Venkataramanan Signed-off-by: Michal Maloszewski Tested-by: Konrad Jankowski Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/iavf/iavf_ethtool.c | 43 +++++++++++++++++++------- 1 file changed, 32 insertions(+), 11 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/iavf/iavf_ethtool.c b/drivers/net/ethernet/intel/iavf/iavf_ethtool.c index 0cecaff38d04..461f5237a2f8 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_ethtool.c +++ b/drivers/net/ethernet/intel/iavf/iavf_ethtool.c @@ -615,23 +615,44 @@ static int iavf_set_ringparam(struct net_device *netdev, if ((ring->rx_mini_pending) || (ring->rx_jumbo_pending)) return -EINVAL; - new_tx_count = clamp_t(u32, ring->tx_pending, - IAVF_MIN_TXD, - IAVF_MAX_TXD); - new_tx_count = ALIGN(new_tx_count, IAVF_REQ_DESCRIPTOR_MULTIPLE); + if (ring->tx_pending > IAVF_MAX_TXD || + ring->tx_pending < IAVF_MIN_TXD || + ring->rx_pending > IAVF_MAX_RXD || + ring->rx_pending < IAVF_MIN_RXD) { + netdev_err(netdev, "Descriptors requested (Tx: %d / Rx: %d) out of range [%d-%d] (increment %d)\n", + ring->tx_pending, ring->rx_pending, IAVF_MIN_TXD, + IAVF_MAX_RXD, IAVF_REQ_DESCRIPTOR_MULTIPLE); + return -EINVAL; + } - new_rx_count = clamp_t(u32, ring->rx_pending, - IAVF_MIN_RXD, - IAVF_MAX_RXD); - new_rx_count = ALIGN(new_rx_count, IAVF_REQ_DESCRIPTOR_MULTIPLE); + new_tx_count = ALIGN(ring->tx_pending, IAVF_REQ_DESCRIPTOR_MULTIPLE); + if (new_tx_count != ring->tx_pending) + netdev_info(netdev, "Requested Tx descriptor count rounded up to %d\n", + new_tx_count); + + new_rx_count = ALIGN(ring->rx_pending, IAVF_REQ_DESCRIPTOR_MULTIPLE); + if (new_rx_count != ring->rx_pending) + netdev_info(netdev, "Requested Rx descriptor count rounded up to %d\n", + new_rx_count); /* if nothing to do return success */ if ((new_tx_count == adapter->tx_desc_count) && - (new_rx_count == adapter->rx_desc_count)) + (new_rx_count == adapter->rx_desc_count)) { + netdev_dbg(netdev, "Nothing to change, descriptor count is same as requested\n"); return 0; + } - adapter->tx_desc_count = new_tx_count; - adapter->rx_desc_count = new_rx_count; + if (new_tx_count != adapter->tx_desc_count) { + netdev_dbg(netdev, "Changing Tx descriptor count from %d to %d\n", + adapter->tx_desc_count, new_tx_count); + adapter->tx_desc_count = new_tx_count; + } + + if (new_rx_count != adapter->rx_desc_count) { + netdev_dbg(netdev, "Changing Rx descriptor count from %d to %d\n", + adapter->rx_desc_count, new_rx_count); + adapter->rx_desc_count = new_rx_count; + } if (netif_running(netdev)) { adapter->flags |= IAVF_FLAG_RESET_NEEDED; -- cgit v1.2.3 From 61125b8be85dfbc7e9c7fe1cc6c6d631ab603516 Mon Sep 17 00:00:00 2001 From: Karen Sornek Date: Fri, 14 May 2021 11:43:13 +0200 Subject: i40e: Fix failed opcode appearing if handling messages from VF Fix failed operation code appearing if handling messages from VF. Implemented by waiting for VF appropriate state if request starts handle while VF reset. Without this patch the message handling request while VF is in a reset state ends with error -5 (I40E_ERR_PARAM). Fixes: 5c3c48ac6bf5 ("i40e: implement virtual device interface") Signed-off-by: Grzegorz Szczurek Signed-off-by: Karen Sornek Tested-by: Tony Brelinski Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 70 +++++++++++++++------- drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h | 2 + 2 files changed, 50 insertions(+), 22 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index 80ae264c99ba..f651861442c2 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -1948,6 +1948,32 @@ static int i40e_vc_send_resp_to_vf(struct i40e_vf *vf, return i40e_vc_send_msg_to_vf(vf, opcode, retval, NULL, 0); } +/** + * i40e_sync_vf_state + * @vf: pointer to the VF info + * @state: VF state + * + * Called from a VF message to synchronize the service with a potential + * VF reset state + **/ +static bool i40e_sync_vf_state(struct i40e_vf *vf, enum i40e_vf_states state) +{ + int i; + + /* When handling some messages, it needs VF state to be set. + * It is possible that this flag is cleared during VF reset, + * so there is a need to wait until the end of the reset to + * handle the request message correctly. + */ + for (i = 0; i < I40E_VF_STATE_WAIT_COUNT; i++) { + if (test_bit(state, &vf->vf_states)) + return true; + usleep_range(10000, 20000); + } + + return test_bit(state, &vf->vf_states); +} + /** * i40e_vc_get_version_msg * @vf: pointer to the VF info @@ -2008,7 +2034,7 @@ static int i40e_vc_get_vf_resources_msg(struct i40e_vf *vf, u8 *msg) size_t len = 0; int ret; - if (!test_bit(I40E_VF_STATE_INIT, &vf->vf_states)) { + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_INIT)) { aq_ret = I40E_ERR_PARAM; goto err; } @@ -2131,7 +2157,7 @@ static int i40e_vc_config_promiscuous_mode_msg(struct i40e_vf *vf, u8 *msg) bool allmulti = false; bool alluni = false; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states)) { + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE)) { aq_ret = I40E_ERR_PARAM; goto err_out; } @@ -2219,7 +2245,7 @@ static int i40e_vc_config_queues_msg(struct i40e_vf *vf, u8 *msg) struct i40e_vsi *vsi; u16 num_qps_all = 0; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states)) { + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE)) { aq_ret = I40E_ERR_PARAM; goto error_param; } @@ -2368,7 +2394,7 @@ static int i40e_vc_config_irq_map_msg(struct i40e_vf *vf, u8 *msg) i40e_status aq_ret = 0; int i; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states)) { + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE)) { aq_ret = I40E_ERR_PARAM; goto error_param; } @@ -2540,7 +2566,7 @@ static int i40e_vc_disable_queues_msg(struct i40e_vf *vf, u8 *msg) struct i40e_pf *pf = vf->pf; i40e_status aq_ret = 0; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states)) { + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE)) { aq_ret = I40E_ERR_PARAM; goto error_param; } @@ -2590,7 +2616,7 @@ static int i40e_vc_request_queues_msg(struct i40e_vf *vf, u8 *msg) u8 cur_pairs = vf->num_queue_pairs; struct i40e_pf *pf = vf->pf; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states)) + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE)) return -EINVAL; if (req_pairs > I40E_MAX_VF_QUEUES) { @@ -2635,7 +2661,7 @@ static int i40e_vc_get_stats_msg(struct i40e_vf *vf, u8 *msg) memset(&stats, 0, sizeof(struct i40e_eth_stats)); - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states)) { + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE)) { aq_ret = I40E_ERR_PARAM; goto error_param; } @@ -2752,7 +2778,7 @@ static int i40e_vc_add_mac_addr_msg(struct i40e_vf *vf, u8 *msg) i40e_status ret = 0; int i; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states) || + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE) || !i40e_vc_isvalid_vsi_id(vf, al->vsi_id)) { ret = I40E_ERR_PARAM; goto error_param; @@ -2824,7 +2850,7 @@ static int i40e_vc_del_mac_addr_msg(struct i40e_vf *vf, u8 *msg) i40e_status ret = 0; int i; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states) || + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE) || !i40e_vc_isvalid_vsi_id(vf, al->vsi_id)) { ret = I40E_ERR_PARAM; goto error_param; @@ -2968,7 +2994,7 @@ static int i40e_vc_remove_vlan_msg(struct i40e_vf *vf, u8 *msg) i40e_status aq_ret = 0; int i; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states) || + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE) || !i40e_vc_isvalid_vsi_id(vf, vfl->vsi_id)) { aq_ret = I40E_ERR_PARAM; goto error_param; @@ -3088,9 +3114,9 @@ static int i40e_vc_config_rss_key(struct i40e_vf *vf, u8 *msg) struct i40e_vsi *vsi = NULL; i40e_status aq_ret = 0; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states) || + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE) || !i40e_vc_isvalid_vsi_id(vf, vrk->vsi_id) || - (vrk->key_len != I40E_HKEY_ARRAY_SIZE)) { + vrk->key_len != I40E_HKEY_ARRAY_SIZE) { aq_ret = I40E_ERR_PARAM; goto err; } @@ -3119,9 +3145,9 @@ static int i40e_vc_config_rss_lut(struct i40e_vf *vf, u8 *msg) i40e_status aq_ret = 0; u16 i; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states) || + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE) || !i40e_vc_isvalid_vsi_id(vf, vrl->vsi_id) || - (vrl->lut_entries != I40E_VF_HLUT_ARRAY_SIZE)) { + vrl->lut_entries != I40E_VF_HLUT_ARRAY_SIZE) { aq_ret = I40E_ERR_PARAM; goto err; } @@ -3154,7 +3180,7 @@ static int i40e_vc_get_rss_hena(struct i40e_vf *vf, u8 *msg) i40e_status aq_ret = 0; int len = 0; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states)) { + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE)) { aq_ret = I40E_ERR_PARAM; goto err; } @@ -3190,7 +3216,7 @@ static int i40e_vc_set_rss_hena(struct i40e_vf *vf, u8 *msg) struct i40e_hw *hw = &pf->hw; i40e_status aq_ret = 0; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states)) { + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE)) { aq_ret = I40E_ERR_PARAM; goto err; } @@ -3215,7 +3241,7 @@ static int i40e_vc_enable_vlan_stripping(struct i40e_vf *vf, u8 *msg) i40e_status aq_ret = 0; struct i40e_vsi *vsi; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states)) { + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE)) { aq_ret = I40E_ERR_PARAM; goto err; } @@ -3241,7 +3267,7 @@ static int i40e_vc_disable_vlan_stripping(struct i40e_vf *vf, u8 *msg) i40e_status aq_ret = 0; struct i40e_vsi *vsi; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states)) { + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE)) { aq_ret = I40E_ERR_PARAM; goto err; } @@ -3468,7 +3494,7 @@ static int i40e_vc_del_cloud_filter(struct i40e_vf *vf, u8 *msg) i40e_status aq_ret = 0; int i, ret; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states)) { + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE)) { aq_ret = I40E_ERR_PARAM; goto err; } @@ -3599,7 +3625,7 @@ static int i40e_vc_add_cloud_filter(struct i40e_vf *vf, u8 *msg) i40e_status aq_ret = 0; int i, ret; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states)) { + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE)) { aq_ret = I40E_ERR_PARAM; goto err_out; } @@ -3708,7 +3734,7 @@ static int i40e_vc_add_qch_msg(struct i40e_vf *vf, u8 *msg) i40e_status aq_ret = 0; u64 speed = 0; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states)) { + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE)) { aq_ret = I40E_ERR_PARAM; goto err; } @@ -3824,7 +3850,7 @@ static int i40e_vc_del_qch_msg(struct i40e_vf *vf, u8 *msg) struct i40e_pf *pf = vf->pf; i40e_status aq_ret = 0; - if (!test_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states)) { + if (!i40e_sync_vf_state(vf, I40E_VF_STATE_ACTIVE)) { aq_ret = I40E_ERR_PARAM; goto err; } diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h index 091e32c1bb46..49575a640a84 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h @@ -18,6 +18,8 @@ #define I40E_MAX_VF_PROMISC_FLAGS 3 +#define I40E_VF_STATE_WAIT_COUNT 20 + /* Various queue ctrls */ enum i40e_queue_ctrl { I40E_QUEUE_CTRL_UNKNOWN = 0, -- cgit v1.2.3 From 8aa55ab422d9d0d825ebfb877702ed661e96e682 Mon Sep 17 00:00:00 2001 From: Mateusz Palczewski Date: Fri, 16 Jul 2021 11:33:56 +0200 Subject: i40e: Fix pre-set max number of queues for VF After setting pre-set combined to 16 queues and reserving 16 queues by tc qdisc, pre-set maximum combined queues returned to default value after VF reset being 4 and this generated errors during removing tc. Fixed by removing clear num_req_queues before reset VF. Fixes: e284fc280473 (i40e: Add and delete cloud filter) Signed-off-by: Grzegorz Szczurek Signed-off-by: Mateusz Palczewski Tested-by: Bindushree P Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 5 ----- 1 file changed, 5 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index f651861442c2..2ea4deb8fc44 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -3823,11 +3823,6 @@ static int i40e_vc_add_qch_msg(struct i40e_vf *vf, u8 *msg) /* set this flag only after making sure all inputs are sane */ vf->adq_enabled = true; - /* num_req_queues is set when user changes number of queues via ethtool - * and this causes issue for default VSI(which depends on this variable) - * when ADq is enabled, hence reset it. - */ - vf->num_req_queues = 0; /* reset the VF in order to allocate resources */ i40e_vc_reset_vf(vf, true); -- cgit v1.2.3 From 23ec111bf3549aae37140330c31a16abfc172421 Mon Sep 17 00:00:00 2001 From: Norbert Zulinski Date: Mon, 22 Nov 2021 12:29:05 +0100 Subject: i40e: Fix NULL pointer dereference in i40e_dbg_dump_desc When trying to dump VFs VSI RX/TX descriptors using debugfs there was a crash due to NULL pointer dereference in i40e_dbg_dump_desc. Added a check to i40e_dbg_dump_desc that checks if VSI type is correct for dumping RX/TX descriptors. Fixes: 02e9c290814c ("i40e: debugfs interface") Signed-off-by: Sylwester Dziedziuch Signed-off-by: Norbert Zulinski Signed-off-by: Mateusz Palczewski Tested-by: Gurucharan G Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/i40e/i40e_debugfs.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c index 291e61ac3e44..2c1b1da1220e 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c +++ b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c @@ -553,6 +553,14 @@ static void i40e_dbg_dump_desc(int cnt, int vsi_seid, int ring_id, int desc_n, dev_info(&pf->pdev->dev, "vsi %d not found\n", vsi_seid); return; } + if (vsi->type != I40E_VSI_MAIN && + vsi->type != I40E_VSI_FDIR && + vsi->type != I40E_VSI_VMDQ2) { + dev_info(&pf->pdev->dev, + "vsi %d type %d descriptor rings not available\n", + vsi_seid, vsi->type); + return; + } if (type == RING_TYPE_XDP && !i40e_enabled_xdp_vsi(vsi)) { dev_info(&pf->pdev->dev, "XDP not enabled on VSI %d\n", vsi_seid); return; -- cgit v1.2.3 From f23ab04dd6f703e282bb2d51fe3ae14f4b88a628 Mon Sep 17 00:00:00 2001 From: Yahui Cao Date: Wed, 5 May 2021 14:18:00 -0700 Subject: ice: fix FDIR init missing when reset VF When VF is being reset, ice_reset_vf() will be called and FDIR resource should be released and initialized again. Fixes: 1f7ea1cd6a37 ("ice: Enable FDIR Configure for AVF") Signed-off-by: Yahui Cao Tested-by: Konrad Jankowski Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c index 217ff5e9a6f1..c2431bc9d9ce 100644 --- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c @@ -1617,6 +1617,7 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr) ice_vc_set_default_allowlist(vf); ice_vf_fdir_exit(vf); + ice_vf_fdir_init(vf); /* clean VF control VSI when resetting VFs since it should be * setup only when VF creates its first FDIR rule. */ @@ -1747,6 +1748,7 @@ bool ice_reset_vf(struct ice_vf *vf, bool is_vflr) } ice_vf_fdir_exit(vf); + ice_vf_fdir_init(vf); /* clean VF control VSI when resetting VF since it should be setup * only when VF creates its first FDIR rule. */ -- cgit v1.2.3 From 2657e16d8c52fb6ffc7250b0b6536f93886e32d6 Mon Sep 17 00:00:00 2001 From: Paul Greenwalt Date: Mon, 12 Jul 2021 07:54:25 -0400 Subject: ice: rearm other interrupt cause register after enabling VFs The other interrupt cause register (OICR), global interrupt 0, is disabled when enabling VFs to prevent handling VFLR. If the OICR is not rearmed then the VF cannot communicate with the PF. Rearm the OICR after enabling VFs. Fixes: 916c7fdf5e93 ("ice: Separate VF VSI initialization/creation from reset flow") Signed-off-by: Paul Greenwalt Tested-by: Tony Brelinski Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c index c2431bc9d9ce..6427e7ec93de 100644 --- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c @@ -2023,6 +2023,10 @@ static int ice_ena_vfs(struct ice_pf *pf, u16 num_vfs) if (ret) goto err_unroll_sriov; + /* rearm global interrupts */ + if (test_and_clear_bit(ICE_OICR_INTR_DIS, pf->state)) + ice_irq_dynamic_ena(hw, NULL, NULL); + return 0; err_unroll_sriov: -- cgit v1.2.3 From 6d39ea19b0fb6cc72427c862b32d39f5af468be3 Mon Sep 17 00:00:00 2001 From: Dave Ertman Date: Tue, 12 Oct 2021 13:31:21 -0700 Subject: ice: Fix problems with DSCP QoS implementation The patch that implemented DSCP QoS implementation removed a bandwidth check that was used to check for a specific condition caused by some corner cases. This check should not of been removed. The same patch also added a check for when the DCBx state could be changed in relation to DSCP, but the check was erroneously added nested in a check for CEE mode, which made the check useless. Fix these problems by re-adding the bandwidth check and relocating the DSCP mode check earlier in the function that changes DCBx state in the driver. Fixes: 2a87bd73e50d ("ice: Add DSCP support") Reported-by: kernel test robot Signed-off-by: Dave Ertman Tested-by: Gurucharan G Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_dcb_nl.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/ice/ice_dcb_nl.c b/drivers/net/ethernet/intel/ice/ice_dcb_nl.c index 7fdeb411b6df..3eb01731e496 100644 --- a/drivers/net/ethernet/intel/ice/ice_dcb_nl.c +++ b/drivers/net/ethernet/intel/ice/ice_dcb_nl.c @@ -97,6 +97,9 @@ static int ice_dcbnl_setets(struct net_device *netdev, struct ieee_ets *ets) new_cfg->etscfg.maxtcs = pf->hw.func_caps.common_cap.maxtc; + if (!bwcfg) + new_cfg->etscfg.tcbwtable[0] = 100; + if (!bwrec) new_cfg->etsrec.tcbwtable[0] = 100; @@ -167,15 +170,18 @@ static u8 ice_dcbnl_setdcbx(struct net_device *netdev, u8 mode) if (mode == pf->dcbx_cap) return ICE_DCB_NO_HW_CHG; - pf->dcbx_cap = mode; qos_cfg = &pf->hw.port_info->qos_cfg; - if (mode & DCB_CAP_DCBX_VER_CEE) { - if (qos_cfg->local_dcbx_cfg.pfc_mode == ICE_QOS_MODE_DSCP) - return ICE_DCB_NO_HW_CHG; + + /* DSCP configuration is not DCBx negotiated */ + if (qos_cfg->local_dcbx_cfg.pfc_mode == ICE_QOS_MODE_DSCP) + return ICE_DCB_NO_HW_CHG; + + pf->dcbx_cap = mode; + + if (mode & DCB_CAP_DCBX_VER_CEE) qos_cfg->local_dcbx_cfg.dcbx_mode = ICE_DCBX_MODE_CEE; - } else { + else qos_cfg->local_dcbx_cfg.dcbx_mode = ICE_DCBX_MODE_IEEE; - } dev_info(ice_pf_to_dev(pf), "DCBx mode = 0x%x\n", mode); return ICE_DCB_HW_CHG_RST; -- cgit v1.2.3 From 28dc1b86f8ea9fd6f4c9e0b363db73ecabf84e22 Mon Sep 17 00:00:00 2001 From: Jesse Brandeburg Date: Fri, 22 Oct 2021 17:28:17 -0700 Subject: ice: ignore dropped packets during init If the hardware is constantly receiving unicast or broadcast packets during driver load, the device previously counted many GLV_RDPC (VSI dropped packets) events during init. This causes confusing dropped packet statistics during driver load. The dropped packets counter incrementing does stop once the driver finishes loading. Avoid this problem by baselining our statistics at the end of driver open instead of the end of probe. Fixes: cdedef59deb0 ("ice: Configure VSIs for Tx/Rx") Signed-off-by: Jesse Brandeburg Tested-by: Gurucharan G Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_main.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 4d1fc48c9744..c6d6ce52e2ca 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -5881,6 +5881,9 @@ static int ice_up_complete(struct ice_vsi *vsi) netif_carrier_on(vsi->netdev); } + /* clear this now, and the first stats read will be used as baseline */ + vsi->stat_offsets_loaded = false; + ice_service_task_schedule(pf); return 0; -- cgit v1.2.3 From 0e32ff024035b693a3304b3ffe30fba58e2ab48c Mon Sep 17 00:00:00 2001 From: Michal Swiatkowski Date: Tue, 16 Nov 2021 11:24:26 +0100 Subject: ice: fix choosing UDP header type In tunnels packet there can be two UDP headers: - outer which for hw should be mark as ICE_UDP_OF - inner which for hw should be mark as ICE_UDP_ILOS or as ICE_TCP_IL if inner header is of TCP type In none tunnels packet header can be: - UDP, which for hw should be mark as ICE_UDP_ILOS - TCP, which for hw should be mark as ICE_TCP_IL Change incorrect ICE_UDP_OF for none tunnel packets to ICE_UDP_ILOS. ICE_UDP_OF is incorrect for none tunnel packets and setting it leads to error from hw while adding this kind of recipe. In summary, for tunnel outer port type should always be set to ICE_UDP_OF, for none tunnel outer and tunnel inner it should always be set to ICE_UDP_ILOS. Fixes: 9e300987d4a8 ("ice: VXLAN and Geneve TC support") Signed-off-by: Michal Swiatkowski Tested-by: Sandeep Penigalapati Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_tc_lib.c | 27 ++++++++++----------------- 1 file changed, 10 insertions(+), 17 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/ice/ice_tc_lib.c b/drivers/net/ethernet/intel/ice/ice_tc_lib.c index e5d23feb6701..384439a267ad 100644 --- a/drivers/net/ethernet/intel/ice/ice_tc_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_tc_lib.c @@ -74,21 +74,13 @@ static enum ice_protocol_type ice_proto_type_from_ipv6(bool inner) return inner ? ICE_IPV6_IL : ICE_IPV6_OFOS; } -static enum ice_protocol_type -ice_proto_type_from_l4_port(bool inner, u16 ip_proto) +static enum ice_protocol_type ice_proto_type_from_l4_port(u16 ip_proto) { - if (inner) { - switch (ip_proto) { - case IPPROTO_UDP: - return ICE_UDP_ILOS; - } - } else { - switch (ip_proto) { - case IPPROTO_TCP: - return ICE_TCP_IL; - case IPPROTO_UDP: - return ICE_UDP_OF; - } + switch (ip_proto) { + case IPPROTO_TCP: + return ICE_TCP_IL; + case IPPROTO_UDP: + return ICE_UDP_ILOS; } return 0; @@ -191,8 +183,9 @@ ice_tc_fill_tunnel_outer(u32 flags, struct ice_tc_flower_fltr *fltr, i++; } - if (flags & ICE_TC_FLWR_FIELD_ENC_DEST_L4_PORT) { - list[i].type = ice_proto_type_from_l4_port(false, hdr->l3_key.ip_proto); + if ((flags & ICE_TC_FLWR_FIELD_ENC_DEST_L4_PORT) && + hdr->l3_key.ip_proto == IPPROTO_UDP) { + list[i].type = ICE_UDP_OF; list[i].h_u.l4_hdr.dst_port = hdr->l4_key.dst_port; list[i].m_u.l4_hdr.dst_port = hdr->l4_mask.dst_port; i++; @@ -317,7 +310,7 @@ ice_tc_fill_rules(struct ice_hw *hw, u32 flags, ICE_TC_FLWR_FIELD_SRC_L4_PORT)) { struct ice_tc_l4_hdr *l4_key, *l4_mask; - list[i].type = ice_proto_type_from_l4_port(inner, headers->l3_key.ip_proto); + list[i].type = ice_proto_type_from_l4_port(headers->l3_key.ip_proto); l4_key = &headers->l4_key; l4_mask = &headers->l4_mask; -- cgit v1.2.3 From de6acd1cdd4d38823b7f4adae82e8a7d62993354 Mon Sep 17 00:00:00 2001 From: Michal Swiatkowski Date: Mon, 22 Nov 2021 16:39:25 +0100 Subject: ice: fix adding different tunnels Adding filters with the same values inside for VXLAN and Geneve causes HW error, because it looks exactly the same. To choose between different type of tunnels new recipe is needed. Add storing tunnel types in creating recipes function and start checking it in finding function. Change getting open tunnels function to return port on correct tunnel type. This is needed to copy correct port to dummy packet. Block user from adding enc_dst_port via tc flower, because VXLAN and Geneve filters can be created only with destination port which was previously opened. Fixes: 8b032a55c1bd5 ("ice: low level support for tunnels") Signed-off-by: Michal Swiatkowski Tested-by: Sandeep Penigalapati Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_ethtool_fdir.c | 4 ++-- drivers/net/ethernet/intel/ice/ice_fdir.c | 2 +- drivers/net/ethernet/intel/ice/ice_flex_pipe.c | 7 +++++-- drivers/net/ethernet/intel/ice/ice_flex_pipe.h | 3 ++- drivers/net/ethernet/intel/ice/ice_switch.c | 19 +++++++++++++------ drivers/net/ethernet/intel/ice/ice_tc_lib.c | 3 ++- 6 files changed, 25 insertions(+), 13 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool_fdir.c b/drivers/net/ethernet/intel/ice/ice_ethtool_fdir.c index 38960bcc384c..b6e7f47c8c78 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool_fdir.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool_fdir.c @@ -1268,7 +1268,7 @@ ice_fdir_write_all_fltr(struct ice_pf *pf, struct ice_fdir_fltr *input, bool is_tun = tun == ICE_FD_HW_SEG_TUN; int err; - if (is_tun && !ice_get_open_tunnel_port(&pf->hw, &port_num)) + if (is_tun && !ice_get_open_tunnel_port(&pf->hw, &port_num, TNL_ALL)) continue; err = ice_fdir_write_fltr(pf, input, add, is_tun); if (err) @@ -1652,7 +1652,7 @@ int ice_add_fdir_ethtool(struct ice_vsi *vsi, struct ethtool_rxnfc *cmd) } /* return error if not an update and no available filters */ - fltrs_needed = ice_get_open_tunnel_port(hw, &tunnel_port) ? 2 : 1; + fltrs_needed = ice_get_open_tunnel_port(hw, &tunnel_port, TNL_ALL) ? 2 : 1; if (!ice_fdir_find_fltr_by_idx(hw, fsp->location) && ice_fdir_num_avail_fltr(hw, pf->vsi[vsi->idx]) < fltrs_needed) { dev_err(dev, "Failed to add filter. The maximum number of flow director filters has been reached.\n"); diff --git a/drivers/net/ethernet/intel/ice/ice_fdir.c b/drivers/net/ethernet/intel/ice/ice_fdir.c index cbd8424631e3..4dca009bdd50 100644 --- a/drivers/net/ethernet/intel/ice/ice_fdir.c +++ b/drivers/net/ethernet/intel/ice/ice_fdir.c @@ -924,7 +924,7 @@ ice_fdir_get_gen_prgm_pkt(struct ice_hw *hw, struct ice_fdir_fltr *input, memcpy(pkt, ice_fdir_pkt[idx].pkt, ice_fdir_pkt[idx].pkt_len); loc = pkt; } else { - if (!ice_get_open_tunnel_port(hw, &tnl_port)) + if (!ice_get_open_tunnel_port(hw, &tnl_port, TNL_ALL)) return ICE_ERR_DOES_NOT_EXIST; if (!ice_fdir_pkt[idx].tun_pkt) return ICE_ERR_PARAM; diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c index 23cfcceb1536..6ad1c2559724 100644 --- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c +++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c @@ -1899,9 +1899,11 @@ static struct ice_buf *ice_pkg_buf(struct ice_buf_build *bld) * ice_get_open_tunnel_port - retrieve an open tunnel port * @hw: pointer to the HW structure * @port: returns open port + * @type: type of tunnel, can be TNL_LAST if it doesn't matter */ bool -ice_get_open_tunnel_port(struct ice_hw *hw, u16 *port) +ice_get_open_tunnel_port(struct ice_hw *hw, u16 *port, + enum ice_tunnel_type type) { bool res = false; u16 i; @@ -1909,7 +1911,8 @@ ice_get_open_tunnel_port(struct ice_hw *hw, u16 *port) mutex_lock(&hw->tnl_lock); for (i = 0; i < hw->tnl.count && i < ICE_TUNNEL_MAX_ENTRIES; i++) - if (hw->tnl.tbl[i].valid && hw->tnl.tbl[i].port) { + if (hw->tnl.tbl[i].valid && hw->tnl.tbl[i].port && + (type == TNL_LAST || type == hw->tnl.tbl[i].type)) { *port = hw->tnl.tbl[i].port; res = true; break; diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.h b/drivers/net/ethernet/intel/ice/ice_flex_pipe.h index 344c2637facd..a2863f38fd1f 100644 --- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.h +++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.h @@ -33,7 +33,8 @@ enum ice_status ice_get_sw_fv_list(struct ice_hw *hw, u8 *prot_ids, u16 ids_cnt, unsigned long *bm, struct list_head *fv_list); bool -ice_get_open_tunnel_port(struct ice_hw *hw, u16 *port); +ice_get_open_tunnel_port(struct ice_hw *hw, u16 *port, + enum ice_tunnel_type type); int ice_udp_tunnel_set_port(struct net_device *netdev, unsigned int table, unsigned int idx, struct udp_tunnel_info *ti); int ice_udp_tunnel_unset_port(struct net_device *netdev, unsigned int table, diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c index 793f4a9fc2cd..183d93033890 100644 --- a/drivers/net/ethernet/intel/ice/ice_switch.c +++ b/drivers/net/ethernet/intel/ice/ice_switch.c @@ -3796,10 +3796,13 @@ static struct ice_protocol_entry ice_prot_id_tbl[ICE_PROTOCOL_LAST] = { * ice_find_recp - find a recipe * @hw: pointer to the hardware structure * @lkup_exts: extension sequence to match + * @tun_type: type of recipe tunnel * * Returns index of matching recipe, or ICE_MAX_NUM_RECIPES if not found. */ -static u16 ice_find_recp(struct ice_hw *hw, struct ice_prot_lkup_ext *lkup_exts) +static u16 +ice_find_recp(struct ice_hw *hw, struct ice_prot_lkup_ext *lkup_exts, + enum ice_sw_tunnel_type tun_type) { bool refresh_required = true; struct ice_sw_recipe *recp; @@ -3860,8 +3863,9 @@ static u16 ice_find_recp(struct ice_hw *hw, struct ice_prot_lkup_ext *lkup_exts) } /* If for "i"th recipe the found was never set to false * then it means we found our match + * Also tun type of recipe needs to be checked */ - if (found) + if (found && recp[i].tun_type == tun_type) return i; /* Return the recipe ID */ } } @@ -4651,11 +4655,12 @@ ice_add_adv_recipe(struct ice_hw *hw, struct ice_adv_lkup_elem *lkups, } /* Look for a recipe which matches our requested fv / mask list */ - *rid = ice_find_recp(hw, lkup_exts); + *rid = ice_find_recp(hw, lkup_exts, rinfo->tun_type); if (*rid < ICE_MAX_NUM_RECIPES) /* Success if found a recipe that match the existing criteria */ goto err_unroll; + rm->tun_type = rinfo->tun_type; /* Recipe we need does not exist, add a recipe */ status = ice_add_sw_recipe(hw, rm, profiles); if (status) @@ -4958,11 +4963,13 @@ ice_fill_adv_packet_tun(struct ice_hw *hw, enum ice_sw_tunnel_type tun_type, switch (tun_type) { case ICE_SW_TUN_VXLAN: + if (!ice_get_open_tunnel_port(hw, &open_port, TNL_VXLAN)) + return ICE_ERR_CFG; + break; case ICE_SW_TUN_GENEVE: - if (!ice_get_open_tunnel_port(hw, &open_port)) + if (!ice_get_open_tunnel_port(hw, &open_port, TNL_GENEVE)) return ICE_ERR_CFG; break; - default: /* Nothing needs to be done for this tunnel type */ return 0; @@ -5555,7 +5562,7 @@ ice_rem_adv_rule(struct ice_hw *hw, struct ice_adv_lkup_elem *lkups, if (status) return status; - rid = ice_find_recp(hw, &lkup_exts); + rid = ice_find_recp(hw, &lkup_exts, rinfo->tun_type); /* If did not find a recipe that match the existing criteria */ if (rid == ICE_MAX_NUM_RECIPES) return ICE_ERR_PARAM; diff --git a/drivers/net/ethernet/intel/ice/ice_tc_lib.c b/drivers/net/ethernet/intel/ice/ice_tc_lib.c index 384439a267ad..25cca5c4ae57 100644 --- a/drivers/net/ethernet/intel/ice/ice_tc_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_tc_lib.c @@ -795,7 +795,8 @@ ice_parse_tunnel_attr(struct net_device *dev, struct flow_rule *rule, headers->l3_mask.ttl = match.mask->ttl; } - if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_PORTS)) { + if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_PORTS) && + fltr->tunnel_type != TNL_VXLAN && fltr->tunnel_type != TNL_GENEVE) { struct flow_match_ports match; flow_rule_match_enc_ports(rule, &match); -- cgit v1.2.3 From 1a0f25a52e08b1f67510cabbb44888d2b3c46359 Mon Sep 17 00:00:00 2001 From: Jesse Brandeburg Date: Fri, 12 Nov 2021 17:06:02 -0800 Subject: ice: safer stats processing The driver was zeroing live stats that could be fetched by ndo_get_stats64 at any time. This could result in inconsistent statistics, and the telltale sign was when reading stats frequently from /proc/net/dev, the stats would go backwards. Fix by collecting stats into a local, and delaying when we write to the structure so it's not incremental. Fixes: fcea6f3da546 ("ice: Add stats and ethtool support") Signed-off-by: Jesse Brandeburg Tested-by: Gurucharan G Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_main.c | 29 ++++++++++++++++++----------- 1 file changed, 18 insertions(+), 11 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index c6d6ce52e2ca..73c61cdb036f 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -5930,14 +5930,15 @@ ice_fetch_u64_stats_per_ring(struct u64_stats_sync *syncp, struct ice_q_stats st /** * ice_update_vsi_tx_ring_stats - Update VSI Tx ring stats counters * @vsi: the VSI to be updated + * @vsi_stats: the stats struct to be updated * @rings: rings to work on * @count: number of rings */ static void -ice_update_vsi_tx_ring_stats(struct ice_vsi *vsi, struct ice_tx_ring **rings, - u16 count) +ice_update_vsi_tx_ring_stats(struct ice_vsi *vsi, + struct rtnl_link_stats64 *vsi_stats, + struct ice_tx_ring **rings, u16 count) { - struct rtnl_link_stats64 *vsi_stats = &vsi->net_stats; u16 i; for (i = 0; i < count; i++) { @@ -5961,15 +5962,13 @@ ice_update_vsi_tx_ring_stats(struct ice_vsi *vsi, struct ice_tx_ring **rings, */ static void ice_update_vsi_ring_stats(struct ice_vsi *vsi) { - struct rtnl_link_stats64 *vsi_stats = &vsi->net_stats; + struct rtnl_link_stats64 *vsi_stats; u64 pkts, bytes; int i; - /* reset netdev stats */ - vsi_stats->tx_packets = 0; - vsi_stats->tx_bytes = 0; - vsi_stats->rx_packets = 0; - vsi_stats->rx_bytes = 0; + vsi_stats = kzalloc(sizeof(*vsi_stats), GFP_ATOMIC); + if (!vsi_stats) + return; /* reset non-netdev (extended) stats */ vsi->tx_restart = 0; @@ -5981,7 +5980,8 @@ static void ice_update_vsi_ring_stats(struct ice_vsi *vsi) rcu_read_lock(); /* update Tx rings counters */ - ice_update_vsi_tx_ring_stats(vsi, vsi->tx_rings, vsi->num_txq); + ice_update_vsi_tx_ring_stats(vsi, vsi_stats, vsi->tx_rings, + vsi->num_txq); /* update Rx rings counters */ ice_for_each_rxq(vsi, i) { @@ -5996,10 +5996,17 @@ static void ice_update_vsi_ring_stats(struct ice_vsi *vsi) /* update XDP Tx rings counters */ if (ice_is_xdp_ena_vsi(vsi)) - ice_update_vsi_tx_ring_stats(vsi, vsi->xdp_rings, + ice_update_vsi_tx_ring_stats(vsi, vsi_stats, vsi->xdp_rings, vsi->num_xdp_txq); rcu_read_unlock(); + + vsi->net_stats.tx_packets = vsi_stats->tx_packets; + vsi->net_stats.tx_bytes = vsi_stats->tx_bytes; + vsi->net_stats.rx_packets = vsi_stats->rx_packets; + vsi->net_stats.rx_bytes = vsi_stats->rx_bytes; + + kfree(vsi_stats); } /** -- cgit v1.2.3 From bc2f39a6252ee40d9bfc2743d4437d420aec5f6e Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 10 Nov 2021 11:13:50 +0300 Subject: iavf: missing unlocks in iavf_watchdog_task() This code was re-organized and there some unlocks missing now. Fixes: 898ef1cb1cb2 ("iavf: Combine init and watchdog state machines") Signed-off-by: Dan Carpenter Tested-by: Konrad Jankowski Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/iavf/iavf_main.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c index cfdbf8c08d18..884a19c51543 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_main.c +++ b/drivers/net/ethernet/intel/iavf/iavf_main.c @@ -2046,6 +2046,7 @@ static void iavf_watchdog_task(struct work_struct *work) } adapter->aq_required = 0; adapter->current_op = VIRTCHNL_OP_UNKNOWN; + mutex_unlock(&adapter->crit_lock); queue_delayed_work(iavf_wq, &adapter->watchdog_task, msecs_to_jiffies(10)); @@ -2076,9 +2077,8 @@ static void iavf_watchdog_task(struct work_struct *work) iavf_detect_recover_hung(&adapter->vsi); break; case __IAVF_REMOVE: - mutex_unlock(&adapter->crit_lock); - return; default: + mutex_unlock(&adapter->crit_lock); return; } -- cgit v1.2.3 From fe523d7c9a8332855376ad5eb1aa301091129ba4 Mon Sep 17 00:00:00 2001 From: Stefan Assmann Date: Wed, 1 Dec 2021 09:14:34 +0100 Subject: iavf: do not override the adapter state in the watchdog task (again) The watchdog task incorrectly changes the state to __IAVF_RESETTING, instead of letting the reset task take care of that. This was already resolved by commit 22c8fd71d3a5 ("iavf: do not override the adapter state in the watchdog task") but the problem was reintroduced by the recent code refactoring in commit 45eebd62999d ("iavf: Refactor iavf state machine tracking"). Fixes: 45eebd62999d ("iavf: Refactor iavf state machine tracking") Signed-off-by: Stefan Assmann Tested-by: Konrad Jankowski Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/iavf/iavf_main.c | 1 - 1 file changed, 1 deletion(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c index 884a19c51543..4e7c04047f91 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_main.c +++ b/drivers/net/ethernet/intel/iavf/iavf_main.c @@ -2085,7 +2085,6 @@ static void iavf_watchdog_task(struct work_struct *work) /* check for hw reset */ reg_val = rd32(hw, IAVF_VF_ARQLEN1) & IAVF_VF_ARQLEN1_ARQENABLE_MASK; if (!reg_val) { - iavf_change_state(adapter, __IAVF_RESETTING); adapter->flags |= IAVF_FLAG_RESET_PENDING; adapter->aq_required = 0; adapter->current_op = VIRTCHNL_OP_UNKNOWN; -- cgit v1.2.3 From 0013881c1145d36bf26165bb70fdd7560a5507a3 Mon Sep 17 00:00:00 2001 From: Karol Kolacinski Date: Thu, 4 Nov 2021 14:52:11 +0100 Subject: ice: Use div64_u64 instead of div_u64 in adjfine Change the division in ice_ptp_adjfine from div_u64 to div64_u64. div_u64 is used when the divisor is 32 bit but in this case incval is 64 bit and it caused incorrect calculations and incval adjustments. Fixes: 06c16d89d2cb ("ice: register 1588 PTP clock device object for E810 devices") Signed-off-by: Karol Kolacinski Tested-by: Gurucharan G Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_ptp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c index bf7247c6f58e..ad7cabe7932f 100644 --- a/drivers/net/ethernet/intel/ice/ice_ptp.c +++ b/drivers/net/ethernet/intel/ice/ice_ptp.c @@ -705,7 +705,7 @@ static int ice_ptp_adjfine(struct ptp_clock_info *info, long scaled_ppm) scaled_ppm = -scaled_ppm; } - while ((u64)scaled_ppm > div_u64(U64_MAX, incval)) { + while ((u64)scaled_ppm > div64_u64(U64_MAX, incval)) { /* handle overflow by scaling down the scaled_ppm and * the divisor, losing some precision */ -- cgit v1.2.3 From 37e738b6fdb14529534dca441e0222313688fde3 Mon Sep 17 00:00:00 2001 From: Karol Kolacinski Date: Tue, 16 Nov 2021 13:07:14 +0100 Subject: ice: Don't put stale timestamps in the skb The driver has to check if it does not accidentally put the timestamp in the SKB before previous timestamp gets overwritten. Timestamp values in the PHY are read only and do not get cleared except at hardware reset or when a new timestamp value is captured. The cached_tstamp field is used to detect the case where a new timestamp has not yet been captured, ensuring that we avoid sending stale timestamp data to the stack. Fixes: ea9b847cda64 ("ice: enable transmit timestamps for E810 devices") Signed-off-by: Karol Kolacinski Tested-by: Gurucharan G Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_ptp.c | 11 ++++------- drivers/net/ethernet/intel/ice/ice_ptp.h | 6 ++++++ 2 files changed, 10 insertions(+), 7 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c index ad7cabe7932f..442b031b0edc 100644 --- a/drivers/net/ethernet/intel/ice/ice_ptp.c +++ b/drivers/net/ethernet/intel/ice/ice_ptp.c @@ -1540,19 +1540,16 @@ static void ice_ptp_tx_tstamp_work(struct kthread_work *work) if (err) continue; - /* Check if the timestamp is valid */ - if (!(raw_tstamp & ICE_PTP_TS_VALID)) + /* Check if the timestamp is invalid or stale */ + if (!(raw_tstamp & ICE_PTP_TS_VALID) || + raw_tstamp == tx->tstamps[idx].cached_tstamp) continue; - /* clear the timestamp register, so that it won't show valid - * again when re-used. - */ - ice_clear_phy_tstamp(hw, tx->quad, phy_idx); - /* The timestamp is valid, so we'll go ahead and clear this * index and then send the timestamp up to the stack. */ spin_lock(&tx->lock); + tx->tstamps[idx].cached_tstamp = raw_tstamp; clear_bit(idx, tx->in_use); skb = tx->tstamps[idx].skb; tx->tstamps[idx].skb = NULL; diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.h b/drivers/net/ethernet/intel/ice/ice_ptp.h index f71ad317d6c8..53c15fc9d996 100644 --- a/drivers/net/ethernet/intel/ice/ice_ptp.h +++ b/drivers/net/ethernet/intel/ice/ice_ptp.h @@ -55,15 +55,21 @@ struct ice_perout_channel { * struct ice_tx_tstamp - Tracking for a single Tx timestamp * @skb: pointer to the SKB for this timestamp request * @start: jiffies when the timestamp was first requested + * @cached_tstamp: last read timestamp * * This structure tracks a single timestamp request. The SKB pointer is * provided when initiating a request. The start time is used to ensure that * we discard old requests that were not fulfilled within a 2 second time * window. + * Timestamp values in the PHY are read only and do not get cleared except at + * hardware reset or when a new timestamp value is captured. The cached_tstamp + * field is used to detect the case where a new timestamp has not yet been + * captured, ensuring that we avoid sending stale timestamp data to the stack. */ struct ice_tx_tstamp { struct sk_buff *skb; unsigned long start; + u64 cached_tstamp; }; /** -- cgit v1.2.3 From 584af82154f56e6b2740160fcc84a2966d969e15 Mon Sep 17 00:00:00 2001 From: Karen Sornek Date: Tue, 31 Aug 2021 13:16:35 +0200 Subject: igb: Fix removal of unicast MAC filters of VFs Move checking condition of VF MAC filter before clearing or adding MAC filter to VF to prevent potential blackout caused by removal of necessary and working VF's MAC filter. Fixes: 1b8b062a99dc ("igb: add VF trust infrastructure") Signed-off-by: Karen Sornek Tested-by: Konrad Jankowski Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/igb/igb_main.c | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c index fd54d3ef890b..b597b8bfb910 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -7648,6 +7648,20 @@ static int igb_set_vf_mac_filter(struct igb_adapter *adapter, const int vf, struct vf_mac_filter *entry = NULL; int ret = 0; + if ((vf_data->flags & IGB_VF_FLAG_PF_SET_MAC) && + !vf_data->trusted) { + dev_warn(&pdev->dev, + "VF %d requested MAC filter but is administratively denied\n", + vf); + return -EINVAL; + } + if (!is_valid_ether_addr(addr)) { + dev_warn(&pdev->dev, + "VF %d attempted to set invalid MAC filter\n", + vf); + return -EINVAL; + } + switch (info) { case E1000_VF_MAC_FILTER_CLR: /* remove all unicast MAC filters related to the current VF */ @@ -7661,20 +7675,6 @@ static int igb_set_vf_mac_filter(struct igb_adapter *adapter, const int vf, } break; case E1000_VF_MAC_FILTER_ADD: - if ((vf_data->flags & IGB_VF_FLAG_PF_SET_MAC) && - !vf_data->trusted) { - dev_warn(&pdev->dev, - "VF %d requested MAC filter but is administratively denied\n", - vf); - return -EINVAL; - } - if (!is_valid_ether_addr(addr)) { - dev_warn(&pdev->dev, - "VF %d attempted to set invalid MAC filter\n", - vf); - return -EINVAL; - } - /* try to find empty slot in the list */ list_for_each(pos, &adapter->vf_macs.l) { entry = list_entry(pos, struct vf_mac_filter, l); -- cgit v1.2.3 From b6d335a60dc624c0d279333b22c737faa765b028 Mon Sep 17 00:00:00 2001 From: Letu Ren Date: Sat, 13 Nov 2021 11:42:34 +0800 Subject: igbvf: fix double free in `igbvf_probe` In `igbvf_probe`, if register_netdev() fails, the program will go to label err_hw_init, and then to label err_ioremap. In free_netdev() which is just below label err_ioremap, there is `list_for_each_entry_safe` and `netif_napi_del` which aims to delete all entries in `dev->napi_list`. The program has added an entry `adapter->rx_ring->napi` which is added by `netif_napi_add` in igbvf_alloc_queues(). However, adapter->rx_ring has been freed below label err_hw_init. So this a UAF. In terms of how to patch the problem, we can refer to igbvf_remove() and delete the entry before `adapter->rx_ring`. The KASAN logs are as follows: [ 35.126075] BUG: KASAN: use-after-free in free_netdev+0x1fd/0x450 [ 35.127170] Read of size 8 at addr ffff88810126d990 by task modprobe/366 [ 35.128360] [ 35.128643] CPU: 1 PID: 366 Comm: modprobe Not tainted 5.15.0-rc2+ #14 [ 35.129789] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.12.0-59-gc9ba5276e321-prebuilt.qemu.org 04/01/2014 [ 35.131749] Call Trace: [ 35.132199] dump_stack_lvl+0x59/0x7b [ 35.132865] print_address_description+0x7c/0x3b0 [ 35.133707] ? free_netdev+0x1fd/0x450 [ 35.134378] __kasan_report+0x160/0x1c0 [ 35.135063] ? free_netdev+0x1fd/0x450 [ 35.135738] kasan_report+0x4b/0x70 [ 35.136367] free_netdev+0x1fd/0x450 [ 35.137006] igbvf_probe+0x121d/0x1a10 [igbvf] [ 35.137808] ? igbvf_vlan_rx_add_vid+0x100/0x100 [igbvf] [ 35.138751] local_pci_probe+0x13c/0x1f0 [ 35.139461] pci_device_probe+0x37e/0x6c0 [ 35.165526] [ 35.165806] Allocated by task 366: [ 35.166414] ____kasan_kmalloc+0xc4/0xf0 [ 35.167117] foo_kmem_cache_alloc_trace+0x3c/0x50 [igbvf] [ 35.168078] igbvf_probe+0x9c5/0x1a10 [igbvf] [ 35.168866] local_pci_probe+0x13c/0x1f0 [ 35.169565] pci_device_probe+0x37e/0x6c0 [ 35.179713] [ 35.179993] Freed by task 366: [ 35.180539] kasan_set_track+0x4c/0x80 [ 35.181211] kasan_set_free_info+0x1f/0x40 [ 35.181942] ____kasan_slab_free+0x103/0x140 [ 35.182703] kfree+0xe3/0x250 [ 35.183239] igbvf_probe+0x1173/0x1a10 [igbvf] [ 35.184040] local_pci_probe+0x13c/0x1f0 Fixes: d4e0fe01a38a0 (igbvf: add new driver to support 82576 virtual functions) Reported-by: Zheyu Ma Signed-off-by: Letu Ren Tested-by: Konrad Jankowski Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/igbvf/netdev.c | 1 + 1 file changed, 1 insertion(+) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/igbvf/netdev.c b/drivers/net/ethernet/intel/igbvf/netdev.c index 74ccd622251a..4d988da68394 100644 --- a/drivers/net/ethernet/intel/igbvf/netdev.c +++ b/drivers/net/ethernet/intel/igbvf/netdev.c @@ -2859,6 +2859,7 @@ static int igbvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent) return 0; err_hw_init: + netif_napi_del(&adapter->rx_ring->napi); kfree(adapter->tx_ring); kfree(adapter->rx_ring); err_sw_init: -- cgit v1.2.3 From 0182d1f3fa640888a2ed7e3f6df2fdb10adee7c8 Mon Sep 17 00:00:00 2001 From: Sasha Neftin Date: Tue, 2 Nov 2021 09:20:06 +0200 Subject: igc: Fix typo in i225 LTR functions The LTR maximum value was incorrectly written using the scale from the LTR minimum value. This would cause incorrect values to be sent, in cases where the initial calculation lead to different min/max scales. Fixes: 707abf069548 ("igc: Add initial LTR support") Suggested-by: Dima Ruinskiy Signed-off-by: Sasha Neftin Tested-by: Nechama Kraus Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/igc/igc_i225.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/igc/igc_i225.c b/drivers/net/ethernet/intel/igc/igc_i225.c index b2ef9fde97b3..b6807e16eea9 100644 --- a/drivers/net/ethernet/intel/igc/igc_i225.c +++ b/drivers/net/ethernet/intel/igc/igc_i225.c @@ -636,7 +636,7 @@ s32 igc_set_ltr_i225(struct igc_hw *hw, bool link) ltrv = rd32(IGC_LTRMAXV); if (ltr_max != (ltrv & IGC_LTRMAXV_LTRV_MASK)) { ltrv = IGC_LTRMAXV_LSNP_REQ | ltr_max | - (scale_min << IGC_LTRMAXV_SCALE_SHIFT); + (scale_max << IGC_LTRMAXV_SCALE_SHIFT); wr32(IGC_LTRMAXV, ltrv); } } -- cgit v1.2.3 From 271225fd57c2f1e0b3f8826df51be6c634affefe Mon Sep 17 00:00:00 2001 From: Robert Schlabbach Date: Tue, 26 Oct 2021 02:24:48 +0200 Subject: ixgbe: Document how to enable NBASE-T support Commit a296d665eae1 ("ixgbe: Add ethtool support to enable 2.5 and 5.0 Gbps support") introduced suppression of the advertisement of NBASE-T speeds by default, according to Todd Fujinaka to accommodate customers with network switches which could not cope with advertised NBASE-T speeds, as posted in the E1000-devel mailing list: https://sourceforge.net/p/e1000/mailman/message/37106269/ However, the suppression was not documented at all, nor was how to enable NBASE-T support. Properly document the NBASE-T suppression and how to enable NBASE-T support. Fixes: a296d665eae1 ("ixgbe: Add ethtool support to enable 2.5 and 5.0 Gbps support") Reported-by: Robert Schlabbach Signed-off-by: Robert Schlabbach Signed-off-by: Tony Nguyen --- .../networking/device_drivers/ethernet/intel/ixgbe.rst | 16 ++++++++++++++++ drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 4 ++++ 2 files changed, 20 insertions(+) (limited to 'drivers/net/ethernet/intel') diff --git a/Documentation/networking/device_drivers/ethernet/intel/ixgbe.rst b/Documentation/networking/device_drivers/ethernet/intel/ixgbe.rst index f1d5233e5e51..0a233b17c664 100644 --- a/Documentation/networking/device_drivers/ethernet/intel/ixgbe.rst +++ b/Documentation/networking/device_drivers/ethernet/intel/ixgbe.rst @@ -440,6 +440,22 @@ NOTE: For 82599-based network connections, if you are enabling jumbo frames in a virtual function (VF), jumbo frames must first be enabled in the physical function (PF). The VF MTU setting cannot be larger than the PF MTU. +NBASE-T Support +--------------- +The ixgbe driver supports NBASE-T on some devices. However, the advertisement +of NBASE-T speeds is suppressed by default, to accommodate broken network +switches which cannot cope with advertised NBASE-T speeds. Use the ethtool +command to enable advertising NBASE-T speeds on devices which support it:: + + ethtool -s eth? advertise 0x1800000001028 + +On Linux systems with INTERFACES(5), this can be specified as a pre-up command +in /etc/network/interfaces so that the interface is always brought up with +NBASE-T support, e.g.:: + + iface eth? inet dhcp + pre-up ethtool -s eth? advertise 0x1800000001028 || true + Generic Receive Offload, aka GRO -------------------------------- The driver supports the in-kernel software implementation of GRO. GRO has diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 0f9f022260d7..45e2ec4d264d 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -5531,6 +5531,10 @@ static int ixgbe_non_sfp_link_config(struct ixgbe_hw *hw) if (!speed && hw->mac.ops.get_link_capabilities) { ret = hw->mac.ops.get_link_capabilities(hw, &speed, &autoneg); + /* remove NBASE-T speeds from default autonegotiation + * to accommodate broken network switches in the field + * which cannot cope with advertised NBASE-T speeds + */ speed &= ~(IXGBE_LINK_SPEED_5GB_FULL | IXGBE_LINK_SPEED_2_5GB_FULL); } -- cgit v1.2.3 From bf0a375055bd1afbbf02a0ef45f7655da7b71317 Mon Sep 17 00:00:00 2001 From: Cyril Novikov Date: Mon, 1 Nov 2021 18:39:36 -0700 Subject: ixgbe: set X550 MDIO speed before talking to PHY The MDIO bus speed must be initialized before talking to the PHY the first time in order to avoid talking to it using a speed that the PHY doesn't support. This fixes HW initialization error -17 (IXGBE_ERR_PHY_ADDR_INVALID) on Denverton CPUs (a.k.a. the Atom C3000 family) on ports with a 10Gb network plugged in. On those devices, HLREG0[MDCSPD] resets to 1, which combined with the 10Gb network results in a 24MHz MDIO speed, which is apparently too fast for the connected PHY. PHY register reads over MDIO bus return garbage, leading to initialization failure. Reproduced with Linux kernel 4.19 and 5.15-rc7. Can be reproduced using the following setup: * Use an Atom C3000 family system with at least one X552 LAN on the SoC * Disable PXE or other BIOS network initialization if possible (the interface must not be initialized before Linux boots) * Connect a live 10Gb Ethernet cable to an X550 port * Power cycle (not reset, doesn't always work) the system and boot Linux * Observe: ixgbe interfaces w/ 10GbE cables plugged in fail with error -17 Fixes: e84db7272798 ("ixgbe: Introduce function to control MDIO speed") Signed-off-by: Cyril Novikov Reviewed-by: Andrew Lunn Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c index 9724ffb16518..e4b50c7781ff 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c @@ -3405,6 +3405,9 @@ static s32 ixgbe_reset_hw_X550em(struct ixgbe_hw *hw) /* flush pending Tx transactions */ ixgbe_clear_tx_pending(hw); + /* set MDIO speed before talking to the PHY in case it's the 1st time */ + ixgbe_set_mdio_speed(hw); + /* PHY ops must be identified and initialized prior to reset */ status = hw->phy.ops.init(hw); if (status == IXGBE_ERR_SFP_NOT_SUPPORTED || -- cgit v1.2.3 From afe8a3ba85ec2a6b6849367e25c06a2f8e0ddd05 Mon Sep 17 00:00:00 2001 From: Maciej Fijalkowski Date: Mon, 13 Dec 2021 16:31:06 +0100 Subject: ice: xsk: return xsk buffers back to pool when cleaning the ring Currently we only NULL the xdp_buff pointer in the internal SW ring but we never give it back to the xsk buffer pool. This means that buffers can be leaked out of the buff pool and never be used again. Add missing xsk_buff_free() call to the routine that is supposed to clean the entries that are left in the ring so that these buffers in the umem can be used by other sockets. Also, only go through the space that is actually left to be cleaned instead of a whole ring. Fixes: 2d4238f55697 ("ice: Add support for AF_XDP") Signed-off-by: Magnus Karlsson Signed-off-by: Maciej Fijalkowski Tested-by: Kiran Bhandare Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_xsk.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c index bb9a80847298..8593717a755e 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.c +++ b/drivers/net/ethernet/intel/ice/ice_xsk.c @@ -811,14 +811,14 @@ bool ice_xsk_any_rx_ring_ena(struct ice_vsi *vsi) */ void ice_xsk_clean_rx_ring(struct ice_rx_ring *rx_ring) { - u16 i; - - for (i = 0; i < rx_ring->count; i++) { - struct xdp_buff **xdp = &rx_ring->xdp_buf[i]; + u16 count_mask = rx_ring->count - 1; + u16 ntc = rx_ring->next_to_clean; + u16 ntu = rx_ring->next_to_use; - if (!xdp) - continue; + for ( ; ntc != ntu; ntc = (ntc + 1) & count_mask) { + struct xdp_buff **xdp = &rx_ring->xdp_buf[ntc]; + xsk_buff_free(*xdp); *xdp = NULL; } } -- cgit v1.2.3 From 617f3e1b588c802517c236087561c6bcb0b4afd6 Mon Sep 17 00:00:00 2001 From: Maciej Fijalkowski Date: Mon, 13 Dec 2021 16:31:07 +0100 Subject: ice: xsk: allocate separate memory for XDP SW ring Currently, the zero-copy data path is reusing the memory region that was initially allocated for an array of struct ice_rx_buf for its own purposes. This is error prone as it is based on the ice_rx_buf struct always being the same size or bigger than what the zero-copy path needs. There can also be old values present in that array giving rise to errors when the zero-copy path uses it. Fix this by freeing the ice_rx_buf region and allocating a new array for the zero-copy path that has the right length and is initialized to zero. Fixes: 57f7f8b6bc0b ("ice: Use xdp_buf instead of rx_buf for xsk zero-copy") Signed-off-by: Maciej Fijalkowski Tested-by: Kiran Bhandare Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_base.c | 17 ++++++++++++++++ drivers/net/ethernet/intel/ice/ice_txrx.c | 19 ++++++++++++------ drivers/net/ethernet/intel/ice/ice_xsk.c | 33 ++++++++++++++++--------------- 3 files changed, 47 insertions(+), 22 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c index 1efc635cc0f5..fafe020e46ee 100644 --- a/drivers/net/ethernet/intel/ice/ice_base.c +++ b/drivers/net/ethernet/intel/ice/ice_base.c @@ -6,6 +6,18 @@ #include "ice_lib.h" #include "ice_dcb_lib.h" +static bool ice_alloc_rx_buf_zc(struct ice_rx_ring *rx_ring) +{ + rx_ring->xdp_buf = kcalloc(rx_ring->count, sizeof(*rx_ring->xdp_buf), GFP_KERNEL); + return !!rx_ring->xdp_buf; +} + +static bool ice_alloc_rx_buf(struct ice_rx_ring *rx_ring) +{ + rx_ring->rx_buf = kcalloc(rx_ring->count, sizeof(*rx_ring->rx_buf), GFP_KERNEL); + return !!rx_ring->rx_buf; +} + /** * __ice_vsi_get_qs_contig - Assign a contiguous chunk of queues to VSI * @qs_cfg: gathered variables needed for PF->VSI queues assignment @@ -492,8 +504,11 @@ int ice_vsi_cfg_rxq(struct ice_rx_ring *ring) xdp_rxq_info_reg(&ring->xdp_rxq, ring->netdev, ring->q_index, ring->q_vector->napi.napi_id); + kfree(ring->rx_buf); ring->xsk_pool = ice_xsk_pool(ring); if (ring->xsk_pool) { + if (!ice_alloc_rx_buf_zc(ring)) + return -ENOMEM; xdp_rxq_info_unreg_mem_model(&ring->xdp_rxq); ring->rx_buf_len = @@ -508,6 +523,8 @@ int ice_vsi_cfg_rxq(struct ice_rx_ring *ring) dev_info(dev, "Registered XDP mem model MEM_TYPE_XSK_BUFF_POOL on Rx ring %d\n", ring->q_index); } else { + if (!ice_alloc_rx_buf(ring)) + return -ENOMEM; if (!xdp_rxq_info_is_reg(&ring->xdp_rxq)) /* coverity[check_return] */ xdp_rxq_info_reg(&ring->xdp_rxq, diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index bc3ba19dc88f..dccf09eefc75 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -419,7 +419,10 @@ void ice_clean_rx_ring(struct ice_rx_ring *rx_ring) } rx_skip_free: - memset(rx_ring->rx_buf, 0, sizeof(*rx_ring->rx_buf) * rx_ring->count); + if (rx_ring->xsk_pool) + memset(rx_ring->xdp_buf, 0, array_size(rx_ring->count, sizeof(*rx_ring->xdp_buf))); + else + memset(rx_ring->rx_buf, 0, array_size(rx_ring->count, sizeof(*rx_ring->rx_buf))); /* Zero out the descriptor ring */ size = ALIGN(rx_ring->count * sizeof(union ice_32byte_rx_desc), @@ -446,8 +449,13 @@ void ice_free_rx_ring(struct ice_rx_ring *rx_ring) if (xdp_rxq_info_is_reg(&rx_ring->xdp_rxq)) xdp_rxq_info_unreg(&rx_ring->xdp_rxq); rx_ring->xdp_prog = NULL; - devm_kfree(rx_ring->dev, rx_ring->rx_buf); - rx_ring->rx_buf = NULL; + if (rx_ring->xsk_pool) { + kfree(rx_ring->xdp_buf); + rx_ring->xdp_buf = NULL; + } else { + kfree(rx_ring->rx_buf); + rx_ring->rx_buf = NULL; + } if (rx_ring->desc) { size = ALIGN(rx_ring->count * sizeof(union ice_32byte_rx_desc), @@ -475,8 +483,7 @@ int ice_setup_rx_ring(struct ice_rx_ring *rx_ring) /* warn if we are about to overwrite the pointer */ WARN_ON(rx_ring->rx_buf); rx_ring->rx_buf = - devm_kcalloc(dev, sizeof(*rx_ring->rx_buf), rx_ring->count, - GFP_KERNEL); + kcalloc(rx_ring->count, sizeof(*rx_ring->rx_buf), GFP_KERNEL); if (!rx_ring->rx_buf) return -ENOMEM; @@ -505,7 +512,7 @@ int ice_setup_rx_ring(struct ice_rx_ring *rx_ring) return 0; err: - devm_kfree(dev, rx_ring->rx_buf); + kfree(rx_ring->rx_buf); rx_ring->rx_buf = NULL; return -ENOMEM; } diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c index 8593717a755e..c124229d98fe 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.c +++ b/drivers/net/ethernet/intel/ice/ice_xsk.c @@ -12,6 +12,11 @@ #include "ice_txrx_lib.h" #include "ice_lib.h" +static struct xdp_buff **ice_xdp_buf(struct ice_rx_ring *rx_ring, u32 idx) +{ + return &rx_ring->xdp_buf[idx]; +} + /** * ice_qp_reset_stats - Resets all stats for rings of given index * @vsi: VSI that contains rings of interest @@ -372,7 +377,7 @@ bool ice_alloc_rx_bufs_zc(struct ice_rx_ring *rx_ring, u16 count) dma_addr_t dma; rx_desc = ICE_RX_DESC(rx_ring, ntu); - xdp = &rx_ring->xdp_buf[ntu]; + xdp = ice_xdp_buf(rx_ring, ntu); nb_buffs = min_t(u16, count, rx_ring->count - ntu); nb_buffs = xsk_buff_alloc_batch(rx_ring->xsk_pool, xdp, nb_buffs); @@ -419,19 +424,18 @@ static void ice_bump_ntc(struct ice_rx_ring *rx_ring) /** * ice_construct_skb_zc - Create an sk_buff from zero-copy buffer * @rx_ring: Rx ring - * @xdp_arr: Pointer to the SW ring of xdp_buff pointers + * @xdp: Pointer to XDP buffer * * This function allocates a new skb from a zero-copy Rx buffer. * * Returns the skb on success, NULL on failure. */ static struct sk_buff * -ice_construct_skb_zc(struct ice_rx_ring *rx_ring, struct xdp_buff **xdp_arr) +ice_construct_skb_zc(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp) { - struct xdp_buff *xdp = *xdp_arr; + unsigned int datasize_hard = xdp->data_end - xdp->data_hard_start; unsigned int metasize = xdp->data - xdp->data_meta; unsigned int datasize = xdp->data_end - xdp->data; - unsigned int datasize_hard = xdp->data_end - xdp->data_hard_start; struct sk_buff *skb; skb = __napi_alloc_skb(&rx_ring->q_vector->napi, datasize_hard, @@ -445,7 +449,6 @@ ice_construct_skb_zc(struct ice_rx_ring *rx_ring, struct xdp_buff **xdp_arr) skb_metadata_set(skb, metasize); xsk_buff_free(xdp); - *xdp_arr = NULL; return skb; } @@ -522,7 +525,7 @@ int ice_clean_rx_irq_zc(struct ice_rx_ring *rx_ring, int budget) while (likely(total_rx_packets < (unsigned int)budget)) { union ice_32b_rx_flex_desc *rx_desc; unsigned int size, xdp_res = 0; - struct xdp_buff **xdp; + struct xdp_buff *xdp; struct sk_buff *skb; u16 stat_err_bits; u16 vlan_tag = 0; @@ -545,18 +548,17 @@ int ice_clean_rx_irq_zc(struct ice_rx_ring *rx_ring, int budget) if (!size) break; - xdp = &rx_ring->xdp_buf[rx_ring->next_to_clean]; - xsk_buff_set_size(*xdp, size); - xsk_buff_dma_sync_for_cpu(*xdp, rx_ring->xsk_pool); + xdp = *ice_xdp_buf(rx_ring, rx_ring->next_to_clean); + xsk_buff_set_size(xdp, size); + xsk_buff_dma_sync_for_cpu(xdp, rx_ring->xsk_pool); - xdp_res = ice_run_xdp_zc(rx_ring, *xdp, xdp_prog, xdp_ring); + xdp_res = ice_run_xdp_zc(rx_ring, xdp, xdp_prog, xdp_ring); if (xdp_res) { if (xdp_res & (ICE_XDP_TX | ICE_XDP_REDIR)) xdp_xmit |= xdp_res; else - xsk_buff_free(*xdp); + xsk_buff_free(xdp); - *xdp = NULL; total_rx_bytes += size; total_rx_packets++; cleaned_count++; @@ -816,10 +818,9 @@ void ice_xsk_clean_rx_ring(struct ice_rx_ring *rx_ring) u16 ntu = rx_ring->next_to_use; for ( ; ntc != ntu; ntc = (ntc + 1) & count_mask) { - struct xdp_buff **xdp = &rx_ring->xdp_buf[ntc]; + struct xdp_buff *xdp = *ice_xdp_buf(rx_ring, ntc); - xsk_buff_free(*xdp); - *xdp = NULL; + xsk_buff_free(xdp); } } -- cgit v1.2.3 From 0708b6facb4d165ef22bccddf2dc3e1eb9a12d03 Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Mon, 13 Dec 2021 16:31:08 +0100 Subject: ice: remove dead store on XSK hotpath The 'if (ntu == rx_ring->count)' block in ice_alloc_rx_buffers_zc() was previously residing in the loop, but after introducing the batched interface it is used only to wrap-around the NTU descriptor, thus no more need to assign 'xdp'. Fixes: db804cfc21e9 ("ice: Use the xsk batched rx allocation interface") Signed-off-by: Alexander Lobakin Acked-by: Maciej Fijalkowski Tested-by: Kiran Bhandare Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_xsk.c | 1 - 1 file changed, 1 deletion(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c index c124229d98fe..27f5f64dcbd6 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.c +++ b/drivers/net/ethernet/intel/ice/ice_xsk.c @@ -397,7 +397,6 @@ bool ice_alloc_rx_bufs_zc(struct ice_rx_ring *rx_ring, u16 count) ntu += nb_buffs; if (ntu == rx_ring->count) { rx_desc = ICE_RX_DESC(rx_ring, 0); - xdp = rx_ring->xdp_buf; ntu = 0; } -- cgit v1.2.3 From 8b51a13c37c24c08e488bd58303cb437814f4454 Mon Sep 17 00:00:00 2001 From: Maciej Fijalkowski Date: Mon, 13 Dec 2021 16:31:09 +0100 Subject: ice: xsk: do not clear status_error0 for ntu + nb_buffs descriptor The descriptor that ntu is pointing at when we exit ice_alloc_rx_bufs_zc() should not have its corresponding DD bit cleared as descriptor is not allocated in there and it is not valid for HW usage. The allocation routine at the entry will fill the descriptor that ntu points to after it was set to ntu + nb_buffs on previous call. Even the spec says: "The tail pointer should be set to one descriptor beyond the last empty descriptor in host descriptor ring." Therefore, step away from clearing the status_error0 on ntu + nb_buffs descriptor. Fixes: db804cfc21e9 ("ice: Use the xsk batched rx allocation interface") Reported-by: Elza Mathew Signed-off-by: Maciej Fijalkowski Tested-by: Kiran Bhandare Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_xsk.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c index 27f5f64dcbd6..ffa9a160766a 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.c +++ b/drivers/net/ethernet/intel/ice/ice_xsk.c @@ -395,13 +395,9 @@ bool ice_alloc_rx_bufs_zc(struct ice_rx_ring *rx_ring, u16 count) } ntu += nb_buffs; - if (ntu == rx_ring->count) { - rx_desc = ICE_RX_DESC(rx_ring, 0); + if (ntu == rx_ring->count) ntu = 0; - } - /* clear the status bits for the next_to_use descriptor */ - rx_desc->wb.status_error0 = 0; ice_release_rx_desc(rx_ring, ntu); return count == nb_buffs; -- cgit v1.2.3 From 8bea15ab7485863d900982ee6a0ff6f78b339c77 Mon Sep 17 00:00:00 2001 From: Maciej Fijalkowski Date: Mon, 13 Dec 2021 16:31:10 +0100 Subject: ice: xsk: allow empty Rx descriptors on XSK ZC data path Commit ac6f733a7bd5 ("ice: allow empty Rx descriptors") stated that ice HW can produce empty descriptors that are valid and they should be processed. Add this support to xsk ZC path to avoid potential processing problems. Fixes: 2d4238f55697 ("ice: Add support for AF_XDP") Signed-off-by: Maciej Fijalkowski Tested-by: Kiran Bhandare Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_xsk.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c index ffa9a160766a..c1491dc0675d 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.c +++ b/drivers/net/ethernet/intel/ice/ice_xsk.c @@ -538,12 +538,18 @@ int ice_clean_rx_irq_zc(struct ice_rx_ring *rx_ring, int budget) */ dma_rmb(); + xdp = *ice_xdp_buf(rx_ring, rx_ring->next_to_clean); + size = le16_to_cpu(rx_desc->wb.pkt_len) & ICE_RX_FLX_DESC_PKT_LEN_M; - if (!size) - break; + if (!size) { + xdp->data = NULL; + xdp->data_end = NULL; + xdp->data_hard_start = NULL; + xdp->data_meta = NULL; + goto construct_skb; + } - xdp = *ice_xdp_buf(rx_ring, rx_ring->next_to_clean); xsk_buff_set_size(xdp, size); xsk_buff_dma_sync_for_cpu(xdp, rx_ring->xsk_pool); @@ -561,7 +567,7 @@ int ice_clean_rx_irq_zc(struct ice_rx_ring *rx_ring, int budget) ice_bump_ntc(rx_ring); continue; } - +construct_skb: /* XDP_PASS path */ skb = ice_construct_skb_zc(rx_ring, xdp); if (!skb) { -- cgit v1.2.3 From dcbaf72aa4232a7aa5db5e483972a6fe4ba2b41c Mon Sep 17 00:00:00 2001 From: Maciej Fijalkowski Date: Mon, 13 Dec 2021 16:31:11 +0100 Subject: ice: xsk: fix cleaned_count setting Currently cleaned_count is initialized to ICE_DESC_UNUSED(rx_ring) and later on during the Rx processing it is incremented per each frame that driver consumed. This can result in excessive buffers requested from xsk pool based on that value. To address this, just drop cleaned_count and pass ICE_DESC_UNUSED(rx_ring) directly as a function argument to ice_alloc_rx_bufs_zc(). Idea is to ask for buffers as many as consumed. Let us also call ice_alloc_rx_bufs_zc unconditionally at the end of ice_clean_rx_irq_zc. This has been changed in that way for corresponding ice_clean_rx_irq, but not here. Fixes: 2d4238f55697 ("ice: Add support for AF_XDP") Signed-off-by: Maciej Fijalkowski Tested-by: Kiran Bhandare Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_txrx.h | 1 - drivers/net/ethernet/intel/ice/ice_xsk.c | 6 +----- 2 files changed, 1 insertion(+), 6 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h index c56dd1749903..b7b3bd4816f0 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.h +++ b/drivers/net/ethernet/intel/ice/ice_txrx.h @@ -24,7 +24,6 @@ #define ICE_MAX_DATA_PER_TXD_ALIGNED \ (~(ICE_MAX_READ_REQ_SIZE - 1) & ICE_MAX_DATA_PER_TXD) -#define ICE_RX_BUF_WRITE 16 /* Must be power of 2 */ #define ICE_MAX_TXQ_PER_TXQG 128 /* Attempt to maximize the headroom available for incoming frames. We use a 2K diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c index c1491dc0675d..c895351b25e0 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.c +++ b/drivers/net/ethernet/intel/ice/ice_xsk.c @@ -505,7 +505,6 @@ out_failure: int ice_clean_rx_irq_zc(struct ice_rx_ring *rx_ring, int budget) { unsigned int total_rx_bytes = 0, total_rx_packets = 0; - u16 cleaned_count = ICE_DESC_UNUSED(rx_ring); struct ice_tx_ring *xdp_ring; unsigned int xdp_xmit = 0; struct bpf_prog *xdp_prog; @@ -562,7 +561,6 @@ int ice_clean_rx_irq_zc(struct ice_rx_ring *rx_ring, int budget) total_rx_bytes += size; total_rx_packets++; - cleaned_count++; ice_bump_ntc(rx_ring); continue; @@ -575,7 +573,6 @@ construct_skb: break; } - cleaned_count++; ice_bump_ntc(rx_ring); if (eth_skb_pad(skb)) { @@ -597,8 +594,7 @@ construct_skb: ice_receive_skb(rx_ring, skb, vlan_tag); } - if (cleaned_count >= ICE_RX_BUF_WRITE) - failure = !ice_alloc_rx_bufs_zc(rx_ring, cleaned_count); + failure = !ice_alloc_rx_bufs_zc(rx_ring, ICE_DESC_UNUSED(rx_ring)); ice_finalize_xdp_rx(xdp_ring, xdp_xmit); ice_update_rx_ring_stats(rx_ring, total_rx_packets, total_rx_bytes); -- cgit v1.2.3 From ac8c58f5b535d6272324e2b8b4a0454781c9147e Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Mon, 20 Dec 2021 12:18:44 -0800 Subject: igb: fix deadlock caused by taking RTNL in RPM resume path Recent net core changes caused an issue with few Intel drivers (reportedly igb), where taking RTNL in RPM resume path results in a deadlock. See [0] for a bug report. I don't think the core changes are wrong, but taking RTNL in RPM resume path isn't needed. The Intel drivers are the only ones doing this. See [1] for a discussion on the issue. Following patch changes the RPM resume path to not take RTNL. [0] https://bugzilla.kernel.org/show_bug.cgi?id=215129 [1] https://lore.kernel.org/netdev/20211125074949.5f897431@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com/t/ Fixes: bd869245a3dc ("net: core: try to runtime-resume detached device in __dev_open") Fixes: f32a21376573 ("ethtool: runtime-resume netdev parent before ethtool ioctl ops") Tested-by: Martin Stolpe Signed-off-by: Heiner Kallweit Signed-off-by: Tony Nguyen Link: https://lore.kernel.org/r/20211220201844.2714498-1-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/igb/igb_main.c | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c index b597b8bfb910..446894dde182 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -9254,7 +9254,7 @@ static int __maybe_unused igb_suspend(struct device *dev) return __igb_shutdown(to_pci_dev(dev), NULL, 0); } -static int __maybe_unused igb_resume(struct device *dev) +static int __maybe_unused __igb_resume(struct device *dev, bool rpm) { struct pci_dev *pdev = to_pci_dev(dev); struct net_device *netdev = pci_get_drvdata(pdev); @@ -9297,17 +9297,24 @@ static int __maybe_unused igb_resume(struct device *dev) wr32(E1000_WUS, ~0); - rtnl_lock(); + if (!rpm) + rtnl_lock(); if (!err && netif_running(netdev)) err = __igb_open(netdev, true); if (!err) netif_device_attach(netdev); - rtnl_unlock(); + if (!rpm) + rtnl_unlock(); return err; } +static int __maybe_unused igb_resume(struct device *dev) +{ + return __igb_resume(dev, false); +} + static int __maybe_unused igb_runtime_idle(struct device *dev) { struct net_device *netdev = dev_get_drvdata(dev); @@ -9326,7 +9333,7 @@ static int __maybe_unused igb_runtime_suspend(struct device *dev) static int __maybe_unused igb_runtime_resume(struct device *dev) { - return igb_resume(dev); + return __igb_resume(dev, true); } static void igb_shutdown(struct pci_dev *pdev) @@ -9442,7 +9449,7 @@ static pci_ers_result_t igb_io_error_detected(struct pci_dev *pdev, * @pdev: Pointer to PCI device * * Restart the card from scratch, as if from a cold-boot. Implementation - * resembles the first-half of the igb_resume routine. + * resembles the first-half of the __igb_resume routine. **/ static pci_ers_result_t igb_io_slot_reset(struct pci_dev *pdev) { @@ -9482,7 +9489,7 @@ static pci_ers_result_t igb_io_slot_reset(struct pci_dev *pdev) * * This callback is called when the error recovery driver tells us that * its OK to resume normal operation. Implementation resembles the - * second-half of the igb_resume routine. + * second-half of the __igb_resume routine. */ static void igb_io_resume(struct pci_dev *pdev) { -- cgit v1.2.3 From 1e81dcc1ab7de7a789e60042ce82d5a612632599 Mon Sep 17 00:00:00 2001 From: Vinicius Costa Gomes Date: Mon, 13 Dec 2021 16:39:49 -0800 Subject: igc: Do not enable crosstimestamping for i225-V models It was reported that when PCIe PTM is enabled, some lockups could be observed with some integrated i225-V models. While the issue is investigated, we can disable crosstimestamp for those models and see no loss of functionality, because those models don't have any support for time synchronization. Fixes: a90ec8483732 ("igc: Add support for PTP getcrosststamp()") Link: https://lore.kernel.org/all/924175a188159f4e03bd69908a91e606b574139b.camel@gmx.de/ Reported-by: Stefan Dietrich Signed-off-by: Vinicius Costa Gomes Tested-by: Nechama Kraus Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/igc/igc_ptp.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/igc/igc_ptp.c b/drivers/net/ethernet/intel/igc/igc_ptp.c index 30568e3544cd..4f9245aa79a1 100644 --- a/drivers/net/ethernet/intel/igc/igc_ptp.c +++ b/drivers/net/ethernet/intel/igc/igc_ptp.c @@ -768,7 +768,20 @@ int igc_ptp_get_ts_config(struct net_device *netdev, struct ifreq *ifr) */ static bool igc_is_crosststamp_supported(struct igc_adapter *adapter) { - return IS_ENABLED(CONFIG_X86_TSC) ? pcie_ptm_enabled(adapter->pdev) : false; + if (!IS_ENABLED(CONFIG_X86_TSC)) + return false; + + /* FIXME: it was noticed that enabling support for PCIe PTM in + * some i225-V models could cause lockups when bringing the + * interface up/down. There should be no downsides to + * disabling crosstimestamping support for i225-V, as it + * doesn't have any PTP support. That way we gain some time + * while root causing the issue. + */ + if (adapter->pdev->device == IGC_DEV_ID_I225_V) + return false; + + return pcie_ptm_enabled(adapter->pdev); } static struct system_counterval_t igc_device_tstamp_to_system(u64 tstamp) -- cgit v1.2.3 From f85846bbf43de38fb2c89fe7d2a085608c4eb25a Mon Sep 17 00:00:00 2001 From: James McLaughlin Date: Fri, 17 Dec 2021 16:49:33 -0700 Subject: igc: Fix TX timestamp support for non-MSI-X platforms Time synchronization was not properly enabled on non-MSI-X platforms. Fixes: 2c344ae24501 ("igc: Add support for TX timestamping") Signed-off-by: James McLaughlin Reviewed-by: Vinicius Costa Gomes Tested-by: Nechama Kraus Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/igc/igc_main.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c index 8e448288ee26..d28a80a00953 100644 --- a/drivers/net/ethernet/intel/igc/igc_main.c +++ b/drivers/net/ethernet/intel/igc/igc_main.c @@ -5467,6 +5467,9 @@ static irqreturn_t igc_intr_msi(int irq, void *data) mod_timer(&adapter->watchdog_timer, jiffies + 1); } + if (icr & IGC_ICR_TS) + igc_tsync_interrupt(adapter); + napi_schedule(&q_vector->napi); return IRQ_HANDLED; @@ -5510,6 +5513,9 @@ static irqreturn_t igc_intr(int irq, void *data) mod_timer(&adapter->watchdog_timer, jiffies + 1); } + if (icr & IGC_ICR_TS) + igc_tsync_interrupt(adapter); + napi_schedule(&q_vector->napi); return IRQ_HANDLED; -- cgit v1.2.3 From 01cbf50877e602e2376af89e4a51c30bc574c618 Mon Sep 17 00:00:00 2001 From: Mateusz Palczewski Date: Wed, 3 Mar 2021 11:45:33 +0000 Subject: i40e: Fix to not show opcode msg on unsuccessful VF MAC change Hide i40e opcode information sent during response to VF in case when untrusted VF tried to change MAC on the VF interface. This is implemented by adding an additional parameter 'hide' to the response sent to VF function that hides the display of error information, but forwards the error code to VF. Previously it was not possible to send response with some error code to VF without displaying opcode information. Fixes: 5c3c48ac6bf5 ("i40e: implement virtual device interface") Signed-off-by: Grzegorz Szczurek Signed-off-by: Mateusz Palczewski Reviewed-by: Paul M Stillwell Jr Reviewed-by: Aleksandr Loktionov Tested-by: Tony Brelinski Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 40 +++++++++++++++++----- 1 file changed, 32 insertions(+), 8 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index 2ea4deb8fc44..048f1678ab8a 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -1877,17 +1877,19 @@ sriov_configure_out: /***********************virtual channel routines******************/ /** - * i40e_vc_send_msg_to_vf + * i40e_vc_send_msg_to_vf_ex * @vf: pointer to the VF info * @v_opcode: virtual channel opcode * @v_retval: virtual channel return value * @msg: pointer to the msg buffer * @msglen: msg length + * @is_quiet: true for not printing unsuccessful return values, false otherwise * * send msg to VF **/ -static int i40e_vc_send_msg_to_vf(struct i40e_vf *vf, u32 v_opcode, - u32 v_retval, u8 *msg, u16 msglen) +static int i40e_vc_send_msg_to_vf_ex(struct i40e_vf *vf, u32 v_opcode, + u32 v_retval, u8 *msg, u16 msglen, + bool is_quiet) { struct i40e_pf *pf; struct i40e_hw *hw; @@ -1903,7 +1905,7 @@ static int i40e_vc_send_msg_to_vf(struct i40e_vf *vf, u32 v_opcode, abs_vf_id = vf->vf_id + hw->func_caps.vf_base_id; /* single place to detect unsuccessful return values */ - if (v_retval) { + if (v_retval && !is_quiet) { vf->num_invalid_msgs++; dev_info(&pf->pdev->dev, "VF %d failed opcode %d, retval: %d\n", vf->vf_id, v_opcode, v_retval); @@ -1933,6 +1935,23 @@ static int i40e_vc_send_msg_to_vf(struct i40e_vf *vf, u32 v_opcode, return 0; } +/** + * i40e_vc_send_msg_to_vf + * @vf: pointer to the VF info + * @v_opcode: virtual channel opcode + * @v_retval: virtual channel return value + * @msg: pointer to the msg buffer + * @msglen: msg length + * + * send msg to VF + **/ +static int i40e_vc_send_msg_to_vf(struct i40e_vf *vf, u32 v_opcode, + u32 v_retval, u8 *msg, u16 msglen) +{ + return i40e_vc_send_msg_to_vf_ex(vf, v_opcode, v_retval, + msg, msglen, false); +} + /** * i40e_vc_send_resp_to_vf * @vf: pointer to the VF info @@ -2695,6 +2714,7 @@ error_param: * i40e_check_vf_permission * @vf: pointer to the VF info * @al: MAC address list from virtchnl + * @is_quiet: set true for printing msg without opcode info, false otherwise * * Check that the given list of MAC addresses is allowed. Will return -EPERM * if any address in the list is not valid. Checks the following conditions: @@ -2709,13 +2729,15 @@ error_param: * addresses might not be accurate. **/ static inline int i40e_check_vf_permission(struct i40e_vf *vf, - struct virtchnl_ether_addr_list *al) + struct virtchnl_ether_addr_list *al, + bool *is_quiet) { struct i40e_pf *pf = vf->pf; struct i40e_vsi *vsi = pf->vsi[vf->lan_vsi_idx]; int mac2add_cnt = 0; int i; + *is_quiet = false; for (i = 0; i < al->num_elements; i++) { struct i40e_mac_filter *f; u8 *addr = al->list[i].addr; @@ -2739,6 +2761,7 @@ static inline int i40e_check_vf_permission(struct i40e_vf *vf, !ether_addr_equal(addr, vf->default_lan_addr.addr)) { dev_err(&pf->pdev->dev, "VF attempting to override administratively set MAC address, bring down and up the VF interface to resume normal operation\n"); + *is_quiet = true; return -EPERM; } @@ -2775,6 +2798,7 @@ static int i40e_vc_add_mac_addr_msg(struct i40e_vf *vf, u8 *msg) (struct virtchnl_ether_addr_list *)msg; struct i40e_pf *pf = vf->pf; struct i40e_vsi *vsi = NULL; + bool is_quiet = false; i40e_status ret = 0; int i; @@ -2791,7 +2815,7 @@ static int i40e_vc_add_mac_addr_msg(struct i40e_vf *vf, u8 *msg) */ spin_lock_bh(&vsi->mac_filter_hash_lock); - ret = i40e_check_vf_permission(vf, al); + ret = i40e_check_vf_permission(vf, al, &is_quiet); if (ret) { spin_unlock_bh(&vsi->mac_filter_hash_lock); goto error_param; @@ -2829,8 +2853,8 @@ static int i40e_vc_add_mac_addr_msg(struct i40e_vf *vf, u8 *msg) error_param: /* send the response to the VF */ - return i40e_vc_send_resp_to_vf(vf, VIRTCHNL_OP_ADD_ETH_ADDR, - ret); + return i40e_vc_send_msg_to_vf_ex(vf, VIRTCHNL_OP_ADD_ETH_ADDR, + ret, NULL, 0, is_quiet); } /** -- cgit v1.2.3 From 3116f59c12bd24c513194cd3acb3ec1f7d468954 Mon Sep 17 00:00:00 2001 From: Di Zhu Date: Mon, 29 Nov 2021 19:52:01 +0600 Subject: i40e: fix use-after-free in i40e_sync_filters_subtask() Using ifconfig command to delete the ipv6 address will cause the i40e network card driver to delete its internal mac_filter and i40e_service_task kernel thread will concurrently access the mac_filter. These two processes are not protected by lock so causing the following use-after-free problems. print_address_description+0x70/0x360 ? vprintk_func+0x5e/0xf0 kasan_report+0x1b2/0x330 i40e_sync_vsi_filters+0x4f0/0x1850 [i40e] i40e_sync_filters_subtask+0xe3/0x130 [i40e] i40e_service_task+0x195/0x24c0 [i40e] process_one_work+0x3f5/0x7d0 worker_thread+0x61/0x6c0 ? process_one_work+0x7d0/0x7d0 kthread+0x1c3/0x1f0 ? kthread_park+0xc0/0xc0 ret_from_fork+0x35/0x40 Allocated by task 2279810: kasan_kmalloc+0xa0/0xd0 kmem_cache_alloc_trace+0xf3/0x1e0 i40e_add_filter+0x127/0x2b0 [i40e] i40e_add_mac_filter+0x156/0x190 [i40e] i40e_addr_sync+0x2d/0x40 [i40e] __hw_addr_sync_dev+0x154/0x210 i40e_set_rx_mode+0x6d/0xf0 [i40e] __dev_set_rx_mode+0xfb/0x1f0 __dev_mc_add+0x6c/0x90 igmp6_group_added+0x214/0x230 __ipv6_dev_mc_inc+0x338/0x4f0 addrconf_join_solict.part.7+0xa2/0xd0 addrconf_dad_work+0x500/0x980 process_one_work+0x3f5/0x7d0 worker_thread+0x61/0x6c0 kthread+0x1c3/0x1f0 ret_from_fork+0x35/0x40 Freed by task 2547073: __kasan_slab_free+0x130/0x180 kfree+0x90/0x1b0 __i40e_del_filter+0xa3/0xf0 [i40e] i40e_del_mac_filter+0xf3/0x130 [i40e] i40e_addr_unsync+0x85/0xa0 [i40e] __hw_addr_sync_dev+0x9d/0x210 i40e_set_rx_mode+0x6d/0xf0 [i40e] __dev_set_rx_mode+0xfb/0x1f0 __dev_mc_del+0x69/0x80 igmp6_group_dropped+0x279/0x510 __ipv6_dev_mc_dec+0x174/0x220 addrconf_leave_solict.part.8+0xa2/0xd0 __ipv6_ifa_notify+0x4cd/0x570 ipv6_ifa_notify+0x58/0x80 ipv6_del_addr+0x259/0x4a0 inet6_addr_del+0x188/0x260 addrconf_del_ifaddr+0xcc/0x130 inet6_ioctl+0x152/0x190 sock_do_ioctl+0xd8/0x2b0 sock_ioctl+0x2e5/0x4c0 do_vfs_ioctl+0x14e/0xa80 ksys_ioctl+0x7c/0xa0 __x64_sys_ioctl+0x42/0x50 do_syscall_64+0x98/0x2c0 entry_SYSCALL_64_after_hwframe+0x65/0xca Fixes: 41c445ff0f48 ("i40e: main driver core") Signed-off-by: Di Zhu Signed-off-by: Rui Zhang Tested-by: Gurucharan G Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/i40e/i40e_main.c | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index e118cf9265c7..e0c4d6113c02 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -99,6 +99,24 @@ MODULE_LICENSE("GPL v2"); static struct workqueue_struct *i40e_wq; +static void netdev_hw_addr_refcnt(struct i40e_mac_filter *f, + struct net_device *netdev, int delta) +{ + struct netdev_hw_addr *ha; + + if (!f || !netdev) + return; + + netdev_for_each_mc_addr(ha, netdev) { + if (ether_addr_equal(ha->addr, f->macaddr)) { + ha->refcount += delta; + if (ha->refcount <= 0) + ha->refcount = 1; + break; + } + } +} + /** * i40e_allocate_dma_mem_d - OS specific memory alloc for shared code * @hw: pointer to the HW structure @@ -2036,6 +2054,7 @@ static void i40e_undo_add_filter_entries(struct i40e_vsi *vsi, hlist_for_each_entry_safe(new, h, from, hlist) { /* We can simply free the wrapper structure */ hlist_del(&new->hlist); + netdev_hw_addr_refcnt(new->f, vsi->netdev, -1); kfree(new); } } @@ -2383,6 +2402,10 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) &tmp_add_list, &tmp_del_list, vlan_filters); + + hlist_for_each_entry(new, &tmp_add_list, hlist) + netdev_hw_addr_refcnt(new->f, vsi->netdev, 1); + if (retval) goto err_no_memory_locked; @@ -2515,6 +2538,7 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) if (new->f->state == I40E_FILTER_NEW) new->f->state = new->state; hlist_del(&new->hlist); + netdev_hw_addr_refcnt(new->f, vsi->netdev, -1); kfree(new); } spin_unlock_bh(&vsi->mac_filter_hash_lock); -- cgit v1.2.3 From 40feded8a247f95957a0de9abd100085fb320a2f Mon Sep 17 00:00:00 2001 From: Mateusz Palczewski Date: Thu, 9 Dec 2021 11:04:35 +0100 Subject: i40e: Fix for displaying message regarding NVM version When loading the i40e driver, it prints a message like: 'The driver for the device detected a newer version of the NVM image v1.x than expected v1.y. Please install the most recent version of the network driver.' This is misleading as the driver is working as expected. Fix that by removing the second part of message and changing it from dev_info to dev_dbg. Fixes: 4fb29bddb57f ("i40e: The driver now prints the API version in error message") Signed-off-by: Mateusz Palczewski Tested-by: Gurucharan G Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/i40e/i40e_main.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index e0c4d6113c02..17c3f6d69740 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -15475,8 +15475,8 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (hw->aq.api_maj_ver == I40E_FW_API_VERSION_MAJOR && hw->aq.api_min_ver > I40E_FW_MINOR_VERSION(hw)) - dev_info(&pdev->dev, - "The driver for the device detected a newer version of the NVM image v%u.%u than expected v%u.%u. Please install the most recent version of the network driver.\n", + dev_dbg(&pdev->dev, + "The driver for the device detected a newer version of the NVM image v%u.%u than v%u.%u.\n", hw->aq.api_maj_ver, hw->aq.api_min_ver, I40E_FW_API_VERSION_MAJOR, -- cgit v1.2.3 From e738451d78b2f8a9635d66c6a87f304b4d965f7a Mon Sep 17 00:00:00 2001 From: Jedrzej Jagielski Date: Fri, 17 Dec 2021 14:29:05 +0000 Subject: i40e: Fix incorrect netdev's real number of RX/TX queues There was a wrong queues representation in sysfs during driver's reinitialization in case of online cpus number is less than combined queues. It was caused by stopped NetworkManager, which is responsible for calling vsi_open function during driver's initialization. In specific situation (ex. 12 cpus online) there were 16 queues in /sys/class/net//queues. In case of modifying queues with value higher, than number of online cpus, then it caused write errors and other errors. Add updating of sysfs's queues representation during driver initialization. Fixes: 41c445ff0f48 ("i40e: main driver core") Signed-off-by: Lukasz Cieplicki Signed-off-by: Jedrzej Jagielski Tested-by: Gurucharan G Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/i40e/i40e_main.c | 32 ++++++++++++++++++++++------- 1 file changed, 25 insertions(+), 7 deletions(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 17c3f6d69740..61afc220fc6c 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -8740,6 +8740,27 @@ int i40e_open(struct net_device *netdev) return 0; } +/** + * i40e_netif_set_realnum_tx_rx_queues - Update number of tx/rx queues + * @vsi: vsi structure + * + * This updates netdev's number of tx/rx queues + * + * Returns status of setting tx/rx queues + **/ +static int i40e_netif_set_realnum_tx_rx_queues(struct i40e_vsi *vsi) +{ + int ret; + + ret = netif_set_real_num_rx_queues(vsi->netdev, + vsi->num_queue_pairs); + if (ret) + return ret; + + return netif_set_real_num_tx_queues(vsi->netdev, + vsi->num_queue_pairs); +} + /** * i40e_vsi_open - * @vsi: the VSI to open @@ -8776,13 +8797,7 @@ int i40e_vsi_open(struct i40e_vsi *vsi) goto err_setup_rx; /* Notify the stack of the actual queue counts. */ - err = netif_set_real_num_tx_queues(vsi->netdev, - vsi->num_queue_pairs); - if (err) - goto err_set_queues; - - err = netif_set_real_num_rx_queues(vsi->netdev, - vsi->num_queue_pairs); + err = i40e_netif_set_realnum_tx_rx_queues(vsi); if (err) goto err_set_queues; @@ -14173,6 +14188,9 @@ struct i40e_vsi *i40e_vsi_setup(struct i40e_pf *pf, u8 type, case I40E_VSI_MAIN: case I40E_VSI_VMDQ2: ret = i40e_config_netdev(vsi); + if (ret) + goto err_netdev; + ret = i40e_netif_set_realnum_tx_rx_queues(vsi); if (ret) goto err_netdev; ret = register_netdev(vsi->netdev); -- cgit v1.2.3 From b712941c8085e638bb92456e866ed3de4404e3d5 Mon Sep 17 00:00:00 2001 From: Karen Sornek Date: Wed, 1 Sep 2021 09:21:46 +0200 Subject: iavf: Fix limit of total number of queues to active queues of VF In the absence of this validation, if the user requests to configure queues more than the enabled queues, it results in sending the requested number of queues to the kernel stack (due to the asynchronous nature of VF response), in which case the stack might pick a queue to transmit that is not enabled and result in Tx hang. Fix this bug by limiting the total number of queues allocated for VF to active queues of VF. Fixes: d5b33d024496 ("i40evf: add ndo_setup_tc callback to i40evf") Signed-off-by: Ashwin Vijayavel Signed-off-by: Karen Sornek Tested-by: Konrad Jankowski Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/iavf/iavf_main.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'drivers/net/ethernet/intel') diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c index 4e7c04047f91..e4439b095533 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_main.c +++ b/drivers/net/ethernet/intel/iavf/iavf_main.c @@ -2708,8 +2708,11 @@ static int iavf_validate_ch_config(struct iavf_adapter *adapter, total_max_rate += tx_rate; num_qps += mqprio_qopt->qopt.count[i]; } - if (num_qps > IAVF_MAX_REQ_QUEUES) + if (num_qps > adapter->num_active_queues) { + dev_err(&adapter->pdev->dev, + "Cannot support requested number of queues\n"); return -EINVAL; + } ret = iavf_validate_tx_bandwidth(adapter, total_max_rate); return ret; -- cgit v1.2.3