From 2e22a75c55c1e70f36e21d4fa4e165af7f356aa9 Mon Sep 17 00:00:00 2001 From: Emil Tantilov Date: Wed, 5 Jul 2017 15:57:30 -0700 Subject: [PATCH 01/10] ixgbe: Clear SWFW_SYNC register during init Added clearing of SW resource bits in the SW/FW synchronization register to ixgbe_init_swfw_sync_X540(). Updated ixgbe_acquire_swfw_sync_X540 SW Manageability host interface resource bit error case to match the error handling of the other SW resource bits. Which is to release the SW resource bits if SW times out while attempting to acquire the resource. This allows the driver to load in cases where the semaphore bits could be stuck after a reset or a crash. Signed-off-by: Emil Tantilov Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ixgbe/ixgbe_x540.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_x540.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_x540.c index 6ea0d6a5fb90..b8c5fd2a2115 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_x540.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_x540.c @@ -619,12 +619,6 @@ s32 ixgbe_acquire_swfw_sync_X540(struct ixgbe_hw *hw, u32 mask) usleep_range(5000, 10000); } - /* Failed to get SW only semaphore */ - if (swmask == IXGBE_GSSR_SW_MNG_SM) { - hw_dbg(hw, "Failed to get SW only semaphore\n"); - return IXGBE_ERR_SWFW_SYNC; - } - /* If the resource is not released by the FW/HW the SW can assume that * the FW/HW malfunctions. In that case the SW should set the SW bit(s) * of the requested resource(s) while ignoring the corresponding FW/HW @@ -647,7 +641,8 @@ s32 ixgbe_acquire_swfw_sync_X540(struct ixgbe_hw *hw, u32 mask) */ if (swfw_sync & swmask) { u32 rmask = IXGBE_GSSR_EEP_SM | IXGBE_GSSR_PHY0_SM | - IXGBE_GSSR_PHY1_SM | IXGBE_GSSR_MAC_CSR_SM; + IXGBE_GSSR_PHY1_SM | IXGBE_GSSR_MAC_CSR_SM | + IXGBE_GSSR_SW_MNG_SM; if (swi2c_mask) rmask |= IXGBE_GSSR_I2C_MASK; @@ -763,6 +758,8 @@ static void ixgbe_release_swfw_sync_semaphore(struct ixgbe_hw *hw) **/ void ixgbe_init_swfw_sync_X540(struct ixgbe_hw *hw) { + u32 rmask; + /* First try to grab the semaphore but we don't need to bother * looking to see whether we got the lock or not since we do * the same thing regardless of whether we got the lock or not. @@ -771,6 +768,14 @@ void ixgbe_init_swfw_sync_X540(struct ixgbe_hw *hw) */ ixgbe_get_swfw_sync_semaphore(hw); ixgbe_release_swfw_sync_semaphore(hw); + + /* Acquire and release all software resources. */ + rmask = IXGBE_GSSR_EEP_SM | IXGBE_GSSR_PHY0_SM | + IXGBE_GSSR_PHY1_SM | IXGBE_GSSR_MAC_CSR_SM | + IXGBE_GSSR_SW_MNG_SM | IXGBE_GSSR_I2C_MASK; + + ixgbe_acquire_swfw_sync_X540(hw, rmask); + ixgbe_release_swfw_sync_X540(hw, rmask); } /** From 2e033eace7557a4b67ba1cb2746bd87d6ca2620b Mon Sep 17 00:00:00 2001 From: Bhumika Goyal Date: Mon, 17 Jul 2017 20:31:18 +0530 Subject: [PATCH 02/10] ixgbe: declare ixgbe_mac_operations structures as const Declare ixgbe_mac_operations structures as const as they are only stored in the mac_ops field of ixgbe_info structure. This field is of type const and therefore ixgbe_mac_operations structure can be made const too. Signed-off-by: Bhumika Goyal Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c index 19fbb2f28ea4..933c5070f1b6 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c @@ -3884,7 +3884,7 @@ static const struct ixgbe_mac_operations mac_ops_X550EM_x_fw = { .write_iosf_sb_reg = ixgbe_write_iosf_sb_reg_x550, }; -static struct ixgbe_mac_operations mac_ops_x550em_a = { +static const struct ixgbe_mac_operations mac_ops_x550em_a = { X550_COMMON_MAC .led_on = ixgbe_led_on_t_x550em, .led_off = ixgbe_led_off_t_x550em, @@ -3905,7 +3905,7 @@ static struct ixgbe_mac_operations mac_ops_x550em_a = { .write_iosf_sb_reg = ixgbe_write_iosf_sb_reg_x550a, }; -static struct ixgbe_mac_operations mac_ops_x550em_a_fw = { +static const struct ixgbe_mac_operations mac_ops_x550em_a_fw = { X550_COMMON_MAC .led_on = ixgbe_led_on_generic, .led_off = ixgbe_led_off_generic, From f5a71caa1763cc8fd1b108234689d6a7e4fe9d2f Mon Sep 17 00:00:00 2001 From: Shannon Nelson Date: Tue, 15 Aug 2017 08:59:54 -0700 Subject: [PATCH 03/10] ixgbe: restore normal RSS after last macvlan offload is removed Just like when the last VF is removed, we need to restore normal operations after the last macvlan offload is removed, else we get stuck in single queue operations. To test: ethtool -l eth1 # note the number of queues in use, ~= cpus ethtool -K eth1 l2-fwd-offload on ip link add mv1 link eth1 type macvlan mode bridge ip link set dev mv1 up ip link del mv1 ethtool -l eth1 # are we back to the same # of queues, or stuck on 1? Signed-off-by: Shannon Nelson Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 3942c6208745..d83cc9d34de3 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -9758,6 +9758,17 @@ static void ixgbe_fwd_del(struct net_device *pdev, void *priv) limit = find_last_bit(&adapter->fwd_bitmask, 32); adapter->ring_feature[RING_F_VMDQ].limit = limit + 1; ixgbe_fwd_ring_down(fwd_adapter->netdev, fwd_adapter); + + /* go back to full RSS if we're done with our VMQs */ + if (adapter->ring_feature[RING_F_VMDQ].limit == 1) { + int rss = min_t(int, ixgbe_max_rss_indices(adapter), + num_online_cpus()); + + adapter->flags &= ~IXGBE_FLAG_VMDQ_ENABLED; + adapter->flags &= ~IXGBE_FLAG_SRIOV_ENABLED; + adapter->ring_feature[RING_F_RSS].limit = rss; + } + ixgbe_setup_tc(pdev, netdev_get_num_tc(pdev)); netdev_dbg(pdev, "pool %i:%i queues %i:%i VSI bitmask %lx\n", fwd_adapter->pool, adapter->num_rx_pools, From c69be946d687a99dbc891ebc66539c1c2f082c1d Mon Sep 17 00:00:00 2001 From: Emil Tantilov Date: Fri, 18 Aug 2017 15:48:02 -0700 Subject: [PATCH 04/10] ixgbe: add error checks when initializing the PHY Ignoring errors when attempting to identify the PHY can lead to a crash. Specifically in the case of FW controlled PHYs where the PHY read/write operations are set to NULL. Removed redundant comment. Signed-off-by: Emil Tantilov Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c index 933c5070f1b6..8cea53b62e1b 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c @@ -3192,6 +3192,9 @@ static s32 ixgbe_init_phy_ops_X550em(struct ixgbe_hw *hw) /* Identify the PHY or SFP module */ ret_val = phy->ops.identify(hw); + if (ret_val == IXGBE_ERR_SFP_NOT_SUPPORTED || + ret_val == IXGBE_ERR_PHY_ADDR_INVALID) + return ret_val; /* Setup function pointers based on detected hardware */ ixgbe_init_mac_link_ops_X550em(hw); @@ -3394,9 +3397,10 @@ static s32 ixgbe_reset_hw_X550em(struct ixgbe_hw *hw) ixgbe_clear_tx_pending(hw); /* PHY ops must be identified and initialized prior to reset */ - - /* Identify PHY and related function pointers */ status = hw->phy.ops.init(hw); + if (status == IXGBE_ERR_SFP_NOT_SUPPORTED || + status == IXGBE_ERR_PHY_ADDR_INVALID) + return status; /* start the external PHY */ if (hw->phy.type == ixgbe_phy_x550em_ext_t) { From 761c2a48c70d871b0622dccaa20ccad024101a51 Mon Sep 17 00:00:00 2001 From: Emil Tantilov Date: Tue, 29 Aug 2017 12:21:48 -0700 Subject: [PATCH 05/10] ixgbe: split Tx/Rx ring clearing for ethtool loopback test Commit: fed21bcee7a5 ("ixgbe: Don't bother clearing buffer memory for descriptor rings) exposed some issues with the logic in the current implementation of ixgbe_clean_test_rings() that are being addressed in this patch: - Split the clearing of the Tx and Rx rings in separate loops. Previously both Tx and Rx rings were cleared in a rx_desc->wb.upper.length based loop which could lead to issues if for w/e reason packets were received outside of the frames transmitted for the loopback test. - Add check for IXGBE_TXD_STAT_DD to avoid clearing the rings if the transmits have not comlpeted by the time we enter ixgbe_clean_test_rings() - Exit early on ixgbe_check_lbtest_frame() failure. This change fixes a crash during ethtool diagnostic (ethtool -t). Signed-off-by: Emil Tantilov Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- .../net/ethernet/intel/ixgbe/ixgbe_ethtool.c | 61 ++++++++++++------- 1 file changed, 38 insertions(+), 23 deletions(-) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c index 72c565712a5f..6d89f28cae06 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c @@ -1916,8 +1916,6 @@ static u16 ixgbe_clean_test_rings(struct ixgbe_ring *rx_ring, unsigned int size) { union ixgbe_adv_rx_desc *rx_desc; - struct ixgbe_rx_buffer *rx_buffer; - struct ixgbe_tx_buffer *tx_buffer; u16 rx_ntc, tx_ntc, count = 0; /* initialize next to clean and descriptor values */ @@ -1925,25 +1923,15 @@ static u16 ixgbe_clean_test_rings(struct ixgbe_ring *rx_ring, tx_ntc = tx_ring->next_to_clean; rx_desc = IXGBE_RX_DESC(rx_ring, rx_ntc); - while (rx_desc->wb.upper.length) { - /* check Rx buffer */ - rx_buffer = &rx_ring->rx_buffer_info[rx_ntc]; + while (tx_ntc != tx_ring->next_to_use) { + union ixgbe_adv_tx_desc *tx_desc; + struct ixgbe_tx_buffer *tx_buffer; - /* sync Rx buffer for CPU read */ - dma_sync_single_for_cpu(rx_ring->dev, - rx_buffer->dma, - ixgbe_rx_bufsz(rx_ring), - DMA_FROM_DEVICE); + tx_desc = IXGBE_TX_DESC(tx_ring, tx_ntc); - /* verify contents of skb */ - if (ixgbe_check_lbtest_frame(rx_buffer, size)) - count++; - - /* sync Rx buffer for device write */ - dma_sync_single_for_device(rx_ring->dev, - rx_buffer->dma, - ixgbe_rx_bufsz(rx_ring), - DMA_FROM_DEVICE); + /* if DD is not set transmit has not completed */ + if (!(tx_desc->wb.status & cpu_to_le32(IXGBE_TXD_STAT_DD))) + return count; /* unmap buffer on Tx side */ tx_buffer = &tx_ring->tx_buffer_info[tx_ntc]; @@ -1958,13 +1946,40 @@ static u16 ixgbe_clean_test_rings(struct ixgbe_ring *rx_ring, DMA_TO_DEVICE); dma_unmap_len_set(tx_buffer, len, 0); - /* increment Rx/Tx next to clean counters */ - rx_ntc++; - if (rx_ntc == rx_ring->count) - rx_ntc = 0; + /* increment Tx next to clean counter */ tx_ntc++; if (tx_ntc == tx_ring->count) tx_ntc = 0; + } + + while (rx_desc->wb.upper.length) { + struct ixgbe_rx_buffer *rx_buffer; + + /* check Rx buffer */ + rx_buffer = &rx_ring->rx_buffer_info[rx_ntc]; + + /* sync Rx buffer for CPU read */ + dma_sync_single_for_cpu(rx_ring->dev, + rx_buffer->dma, + ixgbe_rx_bufsz(rx_ring), + DMA_FROM_DEVICE); + + /* verify contents of skb */ + if (ixgbe_check_lbtest_frame(rx_buffer, size)) + count++; + else + break; + + /* sync Rx buffer for device write */ + dma_sync_single_for_device(rx_ring->dev, + rx_buffer->dma, + ixgbe_rx_bufsz(rx_ring), + DMA_FROM_DEVICE); + + /* increment Rx next to clean counter */ + rx_ntc++; + if (rx_ntc == rx_ring->count) + rx_ntc = 0; /* fetch next descriptor */ rx_desc = IXGBE_RX_DESC(rx_ring, rx_ntc); From 86e23494222f358138e3d2c337f57577b0893797 Mon Sep 17 00:00:00 2001 From: Jesper Dangaard Brouer Date: Mon, 4 Sep 2017 20:40:22 +0200 Subject: [PATCH 06/10] ixgbe: add counter for times Rx pages gets allocated, not recycled The ixgbe driver have page recycle scheme based around the RX-ring queue, where a RX page is shared between two packets. Based on the refcnt, the driver can determine if the RX-page is currently only used by a single packet, if so it can then directly refill/recycle the RX-slot by with the opposite "side" of the page. While this is a clever trick, it is hard to determine when this recycling is successful and when it fails. Adding a counter, which is available via ethtool --statistics as 'alloc_rx_page'. Which counts the number of times the recycle fails and the real page allocator is invoked. When interpreting the stats, do remember that every alloc will serve two packets. The counter is collected per rx_ring, but is summed and ethtool exported as 'alloc_rx_page'. It would be relevant to know what rx_ring that cannot keep up, but that can be exported later if someone experience a need for this. Signed-off-by: Jesper Dangaard Brouer Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ixgbe/ixgbe.h | 2 ++ drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c | 1 + drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 4 ++++ 3 files changed, 7 insertions(+) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h index dd5578756ae0..008d0085e01f 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h @@ -275,6 +275,7 @@ struct ixgbe_rx_queue_stats { u64 rsc_count; u64 rsc_flush; u64 non_eop_descs; + u64 alloc_rx_page; u64 alloc_rx_page_failed; u64 alloc_rx_buff_failed; u64 csum_err; @@ -655,6 +656,7 @@ struct ixgbe_adapter { u64 rsc_total_count; u64 rsc_total_flush; u64 non_eop_descs; + u32 alloc_rx_page; u32 alloc_rx_page_failed; u32 alloc_rx_buff_failed; diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c index 6d89f28cae06..de5704c7dd1b 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c @@ -104,6 +104,7 @@ static const struct ixgbe_stats ixgbe_gstrings_stats[] = { {"tx_flow_control_xoff", IXGBE_STAT(stats.lxofftxc)}, {"rx_flow_control_xoff", IXGBE_STAT(stats.lxoffrxc)}, {"rx_csum_offload_errors", IXGBE_STAT(hw_csum_rx_error)}, + {"alloc_rx_page", IXGBE_STAT(alloc_rx_page)}, {"alloc_rx_page_failed", IXGBE_STAT(alloc_rx_page_failed)}, {"alloc_rx_buff_failed", IXGBE_STAT(alloc_rx_buff_failed)}, {"rx_no_dma_resources", IXGBE_STAT(hw_rx_no_dma_resources)}, diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index d83cc9d34de3..211074934d5b 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -1620,6 +1620,7 @@ static bool ixgbe_alloc_mapped_page(struct ixgbe_ring *rx_ring, bi->page = page; bi->page_offset = ixgbe_rx_offset(rx_ring); bi->pagecnt_bias = 1; + rx_ring->rx_stats.alloc_rx_page++; return true; } @@ -6794,6 +6795,7 @@ void ixgbe_update_stats(struct ixgbe_adapter *adapter) u32 i, missed_rx = 0, mpc, bprc, lxon, lxoff, xon_off_tot; u64 non_eop_descs = 0, restart_queue = 0, tx_busy = 0; u64 alloc_rx_page_failed = 0, alloc_rx_buff_failed = 0; + u64 alloc_rx_page = 0; u64 bytes = 0, packets = 0, hw_csum_rx_error = 0; if (test_bit(__IXGBE_DOWN, &adapter->state) || @@ -6814,6 +6816,7 @@ void ixgbe_update_stats(struct ixgbe_adapter *adapter) for (i = 0; i < adapter->num_rx_queues; i++) { struct ixgbe_ring *rx_ring = adapter->rx_ring[i]; non_eop_descs += rx_ring->rx_stats.non_eop_descs; + alloc_rx_page += rx_ring->rx_stats.alloc_rx_page; alloc_rx_page_failed += rx_ring->rx_stats.alloc_rx_page_failed; alloc_rx_buff_failed += rx_ring->rx_stats.alloc_rx_buff_failed; hw_csum_rx_error += rx_ring->rx_stats.csum_err; @@ -6821,6 +6824,7 @@ void ixgbe_update_stats(struct ixgbe_adapter *adapter) packets += rx_ring->stats.packets; } adapter->non_eop_descs = non_eop_descs; + adapter->alloc_rx_page = alloc_rx_page; adapter->alloc_rx_page_failed = alloc_rx_page_failed; adapter->alloc_rx_buff_failed = alloc_rx_buff_failed; adapter->hw_csum_rx_error = hw_csum_rx_error; From dcfd6b839c998bc9838e2a47f44f37afbdf3099c Mon Sep 17 00:00:00 2001 From: Emil Tantilov Date: Mon, 11 Sep 2017 14:21:31 -0700 Subject: [PATCH 07/10] ixgbe: fix use of uninitialized padding This patch is resolving Coverity hits where padding in a structure could be used uninitialized. - Initialize fwd_cmd.pad/2 before ixgbe_calculate_checksum() - Initialize buffer.pad2/3 before ixgbe_hic_unlocked() Signed-off-by: Emil Tantilov Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ixgbe/ixgbe_common.c | 4 ++-- drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c | 2 ++ 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c index 2c19070d2a0b..041940c4bb2b 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c @@ -3800,10 +3800,10 @@ s32 ixgbe_set_fw_drv_ver_generic(struct ixgbe_hw *hw, u8 maj, u8 min, fw_cmd.ver_build = build; fw_cmd.ver_sub = sub; fw_cmd.hdr.checksum = 0; - fw_cmd.hdr.checksum = ixgbe_calculate_checksum((u8 *)&fw_cmd, - (FW_CEM_HDR_LEN + fw_cmd.hdr.buf_len)); fw_cmd.pad = 0; fw_cmd.pad2 = 0; + fw_cmd.hdr.checksum = ixgbe_calculate_checksum((u8 *)&fw_cmd, + (FW_CEM_HDR_LEN + fw_cmd.hdr.buf_len)); for (i = 0; i <= FW_CEM_MAX_RETRIES; i++) { ret_val = ixgbe_host_interface_command(hw, &fw_cmd, diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c index 8cea53b62e1b..cb7da5f9c4da 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c @@ -900,6 +900,8 @@ static s32 ixgbe_read_ee_hostif_buffer_X550(struct ixgbe_hw *hw, /* convert offset from words to bytes */ buffer.address = cpu_to_be32((offset + current_word) * 2); buffer.length = cpu_to_be16(words_to_read * 2); + buffer.pad2 = 0; + buffer.pad3 = 0; status = ixgbe_hic_unlocked(hw, (u32 *)&buffer, sizeof(buffer), IXGBE_HI_COMMAND_TIMEOUT); From c3aec05dfe2cb1bc1fdf3593da82bc987211338a Mon Sep 17 00:00:00 2001 From: Emil Tantilov Date: Mon, 11 Sep 2017 14:21:36 -0700 Subject: [PATCH 08/10] ixgbe: fix the FWSM.PT check in ixgbe_mng_present() Bits other than FWSM.PT can be set in IXGBE_SWFW_MODE_MASK making the previous check invalid. Change the check for MNG present to be only based on FWSM.PT bit. Signed-off-by: Emil Tantilov Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ixgbe/ixgbe_common.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c index 041940c4bb2b..4e5c92dea869 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c @@ -4100,8 +4100,8 @@ bool ixgbe_mng_present(struct ixgbe_hw *hw) return false; fwsm = IXGBE_READ_REG(hw, IXGBE_FWSM(hw)); - fwsm &= IXGBE_FWSM_MODE_MASK; - return fwsm == IXGBE_FWSM_FW_MODE_PT; + + return !!(fwsm & IXGBE_FWSM_FW_MODE_PT); } /** From b4ded8327fea82b53fcec39e0845011246d020f4 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Mon, 25 Sep 2017 14:55:36 -0700 Subject: [PATCH 09/10] ixgbe: Update adaptive ITR algorithm The following change is meant to update the adaptive ITR algorithm to better support the needs of the network. Specifically with this change what I have done is make it so that our ITR algorithm will try to prevent either starving a socket buffer for memory in the case of Tx, or overrunning an Rx socket buffer on receive. In addition a side effect of the calculations used is that we should function better with new features such as XDP which can handle small packets at high rates without needing to lock us into NAPI polling mode. Signed-off-by: Alexander Duyck Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ixgbe/ixgbe.h | 7 + drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c | 11 +- drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 217 +++++++++++++----- 3 files changed, 179 insertions(+), 56 deletions(-) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h index 008d0085e01f..468c3555a629 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h @@ -435,8 +435,15 @@ static inline unsigned int ixgbe_rx_pg_order(struct ixgbe_ring *ring) } #define ixgbe_rx_pg_size(_ring) (PAGE_SIZE << ixgbe_rx_pg_order(_ring)) +#define IXGBE_ITR_ADAPTIVE_MIN_INC 2 +#define IXGBE_ITR_ADAPTIVE_MIN_USECS 10 +#define IXGBE_ITR_ADAPTIVE_MAX_USECS 126 +#define IXGBE_ITR_ADAPTIVE_LATENCY 0x80 +#define IXGBE_ITR_ADAPTIVE_BULK 0x00 + struct ixgbe_ring_container { struct ixgbe_ring *ring; /* pointer to linked list of rings */ + unsigned long next_update; /* jiffies value of last update */ unsigned int total_bytes; /* total bytes processed this int */ unsigned int total_packets; /* total packets processed this int */ u16 work_limit; /* total work allowed per interrupt */ diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c index f1bfae0c41d0..8e2a957aca18 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c @@ -806,6 +806,7 @@ static void ixgbe_add_ring(struct ixgbe_ring *ring, ring->next = head->ring; head->ring = ring; head->count++; + head->next_update = jiffies + 1; } /** @@ -879,8 +880,11 @@ static int ixgbe_alloc_q_vector(struct ixgbe_adapter *adapter, /* initialize work limits */ q_vector->tx.work_limit = adapter->tx_work_limit; - /* initialize pointer to rings */ - ring = q_vector->ring; + /* Initialize setting for adaptive ITR */ + q_vector->tx.itr = IXGBE_ITR_ADAPTIVE_MAX_USECS | + IXGBE_ITR_ADAPTIVE_LATENCY; + q_vector->rx.itr = IXGBE_ITR_ADAPTIVE_MAX_USECS | + IXGBE_ITR_ADAPTIVE_LATENCY; /* intialize ITR */ if (txr_count && !rxr_count) { @@ -897,6 +901,9 @@ static int ixgbe_alloc_q_vector(struct ixgbe_adapter *adapter, q_vector->itr = adapter->rx_itr_setting; } + /* initialize pointer to rings */ + ring = q_vector->ring; + while (txr_count) { /* assign generic ring traits */ ring->dev = &adapter->pdev->dev; diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 211074934d5b..5e2686d106db 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -2540,50 +2540,174 @@ enum latency_range { static void ixgbe_update_itr(struct ixgbe_q_vector *q_vector, struct ixgbe_ring_container *ring_container) { - int bytes = ring_container->total_bytes; - int packets = ring_container->total_packets; - u32 timepassed_us; - u64 bytes_perint; - u8 itr_setting = ring_container->itr; + unsigned int itr = IXGBE_ITR_ADAPTIVE_MIN_USECS | + IXGBE_ITR_ADAPTIVE_LATENCY; + unsigned int avg_wire_size, packets, bytes; + unsigned long next_update = jiffies; - if (packets == 0) - return; - - /* simple throttlerate management - * 0-10MB/s lowest (100000 ints/s) - * 10-20MB/s low (20000 ints/s) - * 20-1249MB/s bulk (12000 ints/s) + /* If we don't have any rings just leave ourselves set for maximum + * possible latency so we take ourselves out of the equation. */ - /* what was last interrupt timeslice? */ - timepassed_us = q_vector->itr >> 2; - if (timepassed_us == 0) + if (!ring_container->ring) return; - bytes_perint = bytes / timepassed_us; /* bytes/usec */ + /* If we didn't update within up to 1 - 2 jiffies we can assume + * that either packets are coming in so slow there hasn't been + * any work, or that there is so much work that NAPI is dealing + * with interrupt moderation and we don't need to do anything. + */ + if (time_after(next_update, ring_container->next_update)) + goto clear_counts; - switch (itr_setting) { - case lowest_latency: - if (bytes_perint > 10) - itr_setting = low_latency; + packets = ring_container->total_packets; + + /* We have no packets to actually measure against. This means + * either one of the other queues on this vector is active or + * we are a Tx queue doing TSO with too high of an interrupt rate. + * + * When this occurs just tick up our delay by the minimum value + * and hope that this extra delay will prevent us from being called + * without any work on our queue. + */ + if (!packets) { + itr = (q_vector->itr >> 2) + IXGBE_ITR_ADAPTIVE_MIN_INC; + if (itr > IXGBE_ITR_ADAPTIVE_MAX_USECS) + itr = IXGBE_ITR_ADAPTIVE_MAX_USECS; + itr += ring_container->itr & IXGBE_ITR_ADAPTIVE_LATENCY; + goto clear_counts; + } + + bytes = ring_container->total_bytes; + + /* If packets are less than 4 or bytes are less than 9000 assume + * insufficient data to use bulk rate limiting approach. We are + * likely latency driven. + */ + if (packets < 4 && bytes < 9000) { + itr = IXGBE_ITR_ADAPTIVE_LATENCY; + goto adjust_by_size; + } + + /* Between 4 and 48 we can assume that our current interrupt delay + * is only slightly too low. As such we should increase it by a small + * fixed amount. + */ + if (packets < 48) { + itr = (q_vector->itr >> 2) + IXGBE_ITR_ADAPTIVE_MIN_INC; + if (itr > IXGBE_ITR_ADAPTIVE_MAX_USECS) + itr = IXGBE_ITR_ADAPTIVE_MAX_USECS; + goto clear_counts; + } + + /* Between 48 and 96 is our "goldilocks" zone where we are working + * out "just right". Just report that our current ITR is good for us. + */ + if (packets < 96) { + itr = q_vector->itr >> 2; + goto clear_counts; + } + + /* If packet count is 96 or greater we are likely looking at a slight + * overrun of the delay we want. Try halving our delay to see if that + * will cut the number of packets in half per interrupt. + */ + if (packets < 256) { + itr = q_vector->itr >> 3; + if (itr < IXGBE_ITR_ADAPTIVE_MIN_USECS) + itr = IXGBE_ITR_ADAPTIVE_MIN_USECS; + goto clear_counts; + } + + /* The paths below assume we are dealing with a bulk ITR since number + * of packets is 256 or greater. We are just going to have to compute + * a value and try to bring the count under control, though for smaller + * packet sizes there isn't much we can do as NAPI polling will likely + * be kicking in sooner rather than later. + */ + itr = IXGBE_ITR_ADAPTIVE_BULK; + +adjust_by_size: + /* If packet counts are 256 or greater we can assume we have a gross + * overestimation of what the rate should be. Instead of trying to fine + * tune it just use the formula below to try and dial in an exact value + * give the current packet size of the frame. + */ + avg_wire_size = bytes / packets; + + /* The following is a crude approximation of: + * wmem_default / (size + overhead) = desired_pkts_per_int + * rate / bits_per_byte / (size + ethernet overhead) = pkt_rate + * (desired_pkt_rate / pkt_rate) * usecs_per_sec = ITR value + * + * Assuming wmem_default is 212992 and overhead is 640 bytes per + * packet, (256 skb, 64 headroom, 320 shared info), we can reduce the + * formula down to + * + * (170 * (size + 24)) / (size + 640) = ITR + * + * We first do some math on the packet size and then finally bitshift + * by 8 after rounding up. We also have to account for PCIe link speed + * difference as ITR scales based on this. + */ + if (avg_wire_size <= 60) { + /* Start at 50k ints/sec */ + avg_wire_size = 5120; + } else if (avg_wire_size <= 316) { + /* 50K ints/sec to 16K ints/sec */ + avg_wire_size *= 40; + avg_wire_size += 2720; + } else if (avg_wire_size <= 1084) { + /* 16K ints/sec to 9.2K ints/sec */ + avg_wire_size *= 15; + avg_wire_size += 11452; + } else if (avg_wire_size <= 1980) { + /* 9.2K ints/sec to 8K ints/sec */ + avg_wire_size *= 5; + avg_wire_size += 22420; + } else { + /* plateau at a limit of 8K ints/sec */ + avg_wire_size = 32256; + } + + /* If we are in low latency mode half our delay which doubles the rate + * to somewhere between 100K to 16K ints/sec + */ + if (itr & IXGBE_ITR_ADAPTIVE_LATENCY) + avg_wire_size >>= 1; + + /* Resultant value is 256 times larger than it needs to be. This + * gives us room to adjust the value as needed to either increase + * or decrease the value based on link speeds of 10G, 2.5G, 1G, etc. + * + * Use addition as we have already recorded the new latency flag + * for the ITR value. + */ + switch (q_vector->adapter->link_speed) { + case IXGBE_LINK_SPEED_10GB_FULL: + case IXGBE_LINK_SPEED_100_FULL: + default: + itr += DIV_ROUND_UP(avg_wire_size, + IXGBE_ITR_ADAPTIVE_MIN_INC * 256) * + IXGBE_ITR_ADAPTIVE_MIN_INC; break; - case low_latency: - if (bytes_perint > 20) - itr_setting = bulk_latency; - else if (bytes_perint <= 10) - itr_setting = lowest_latency; - break; - case bulk_latency: - if (bytes_perint <= 20) - itr_setting = low_latency; + case IXGBE_LINK_SPEED_2_5GB_FULL: + case IXGBE_LINK_SPEED_1GB_FULL: + case IXGBE_LINK_SPEED_10_FULL: + itr += DIV_ROUND_UP(avg_wire_size, + IXGBE_ITR_ADAPTIVE_MIN_INC * 64) * + IXGBE_ITR_ADAPTIVE_MIN_INC; break; } - /* clear work counters since we have the values we need */ +clear_counts: + /* write back value */ + ring_container->itr = itr; + + /* next update should occur within next jiffy */ + ring_container->next_update = next_update + 1; + ring_container->total_bytes = 0; ring_container->total_packets = 0; - - /* write updated itr to ring container */ - ring_container->itr = itr_setting; } /** @@ -2625,34 +2749,19 @@ void ixgbe_write_eitr(struct ixgbe_q_vector *q_vector) static void ixgbe_set_itr(struct ixgbe_q_vector *q_vector) { - u32 new_itr = q_vector->itr; - u8 current_itr; + u32 new_itr; ixgbe_update_itr(q_vector, &q_vector->tx); ixgbe_update_itr(q_vector, &q_vector->rx); - current_itr = max(q_vector->rx.itr, q_vector->tx.itr); + /* use the smallest value of new ITR delay calculations */ + new_itr = min(q_vector->rx.itr, q_vector->tx.itr); - switch (current_itr) { - /* counts and packets in update_itr are dependent on these numbers */ - case lowest_latency: - new_itr = IXGBE_100K_ITR; - break; - case low_latency: - new_itr = IXGBE_20K_ITR; - break; - case bulk_latency: - new_itr = IXGBE_12K_ITR; - break; - default: - break; - } + /* Clear latency flag if set, shift into correct position */ + new_itr &= ~IXGBE_ITR_ADAPTIVE_LATENCY; + new_itr <<= 2; if (new_itr != q_vector->itr) { - /* do an exponential smoothing */ - new_itr = (10 * new_itr * q_vector->itr) / - ((9 * new_itr) + q_vector->itr); - /* save the algorithm value here */ q_vector->itr = new_itr; From b64666ae00327efe53613fc180df6ffbeee1d8d1 Mon Sep 17 00:00:00 2001 From: Emil Tantilov Date: Fri, 29 Sep 2017 10:55:42 -0700 Subject: [PATCH 10/10] ixgbe: fix crash when injecting AER after failed reset In case where AER recovery fails the device is left in a down state. Consecutive AER error injection can lead to a double IRQ free. Signed-off-by: Emil Tantilov Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 5e2686d106db..c6f9da7990c7 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -10861,6 +10861,9 @@ static pci_ers_result_t ixgbe_io_error_detected(struct pci_dev *pdev, if (!test_bit(__IXGBE_SERVICE_INITED, &adapter->state)) return PCI_ERS_RESULT_DISCONNECT; + if (!netif_device_present(netdev)) + return PCI_ERS_RESULT_DISCONNECT; + rtnl_lock(); netif_device_detach(netdev);