From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Kirsher Subject: [net-next 8/8] ixgbe: Improve statistics accuracy for DDP traffic Date: Sun, 16 Sep 2012 21:15:42 -0700 Message-ID: <1347855342-6610-9-git-send-email-jeffrey.t.kirsher@intel.com> References: <1347855342-6610-1-git-send-email-jeffrey.t.kirsher@intel.com> Cc: Mark Rustad , netdev@vger.kernel.org, gospo@redhat.com, sassmann@redhat.com, Jeff Kirsher To: davem@davemloft.net Return-path: Received: from mga09.intel.com ([134.134.136.24]:46001 "EHLO mga09.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750819Ab2IQEPu (ORCPT ); Mon, 17 Sep 2012 00:15:50 -0400 In-Reply-To: <1347855342-6610-1-git-send-email-jeffrey.t.kirsher@intel.com> Sender: netdev-owner@vger.kernel.org List-ID: From: Mark Rustad Noticed that the byte and packet count statistics are under- counting traffic handled by the DDP offload when there is more than one DDP completion processed in a single call to ixgbe_clean_rx_irq. This patch fixes that. I tried to optimize the setting of the rss value so that it only would have to be computed once, and only when there is a DDP completion present. Signed-off-by: Mark Rustad Tested-by: Phil Schmitt Tested-by: Ross Brattain Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 32 +++++++++++++-------------- 1 file changed, 16 insertions(+), 16 deletions(-) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index b3b846b..2dc9d91 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -1785,7 +1785,8 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector, unsigned int total_rx_bytes = 0, total_rx_packets = 0; #ifdef IXGBE_FCOE struct ixgbe_adapter *adapter = q_vector->adapter; - int ddp_bytes = 0; + int ddp_bytes; + unsigned int mss = 0; #endif /* IXGBE_FCOE */ u16 cleaned_count = ixgbe_desc_unused(rx_ring); @@ -1839,6 +1840,20 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector, /* if ddp, not passing to ULD unless for FCP_RSP or error */ if (ixgbe_rx_is_fcoe(rx_ring, rx_desc)) { ddp_bytes = ixgbe_fcoe_ddp(adapter, rx_desc, skb); + /* include DDPed FCoE data */ + if (ddp_bytes > 0) { + if (!mss) { + mss = rx_ring->netdev->mtu - + sizeof(struct fcoe_hdr) - + sizeof(struct fc_frame_header) - + sizeof(struct fcoe_crc_eof); + if (mss > 512) + mss &= ~511; + } + total_rx_bytes += ddp_bytes; + total_rx_packets += DIV_ROUND_UP(ddp_bytes, + mss); + } if (!ddp_bytes) { dev_kfree_skb_any(skb); continue; @@ -1852,21 +1867,6 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector, budget--; } while (likely(budget)); -#ifdef IXGBE_FCOE - /* include DDPed FCoE data */ - if (ddp_bytes > 0) { - unsigned int mss; - - mss = rx_ring->netdev->mtu - sizeof(struct fcoe_hdr) - - sizeof(struct fc_frame_header) - - sizeof(struct fcoe_crc_eof); - if (mss > 512) - mss &= ~511; - total_rx_bytes += ddp_bytes; - total_rx_packets += DIV_ROUND_UP(ddp_bytes, mss); - } - -#endif /* IXGBE_FCOE */ u64_stats_update_begin(&rx_ring->syncp); rx_ring->stats.packets += total_rx_packets; rx_ring->stats.bytes += total_rx_bytes; -- 1.7.11.4