From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Kirsher Subject: [net-next 15/15] igb: Only DMA sync frame length Date: Wed, 29 Jun 2016 14:33:02 -0700 Message-ID: <1467235982-55437-16-git-send-email-jeffrey.t.kirsher@intel.com> References: <1467235982-55437-1-git-send-email-jeffrey.t.kirsher@intel.com> Cc: Andrew Lunn , netdev@vger.kernel.org, nhorman@redhat.com, sassmann@redhat.com, jogreene@redhat.com, Jeff Kirsher To: davem@davemloft.net Return-path: Received: from mga02.intel.com ([134.134.136.20]:3817 "EHLO mga02.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751793AbcF2VdJ (ORCPT ); Wed, 29 Jun 2016 17:33:09 -0400 In-Reply-To: <1467235982-55437-1-git-send-email-jeffrey.t.kirsher@intel.com> Sender: netdev-owner@vger.kernel.org List-ID: From: Andrew Lunn On some platforms, syncing a buffer for DMA is expensive. Rather than sync the whole 2K receive buffer, only synchronise the length of the frame, which will typically be the MTU, or a much smaller TCP ACK. For an IMX6Q, this gives around 6% increased TCP receive performance, which is cache operations bound and reduces CPU load for TCP transmit. Signed-off-by: Andrew Lunn Tested-by: Aaron Brown Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/igb/igb_main.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c index a15f826..9bcba42 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -6856,12 +6856,12 @@ static bool igb_can_reuse_rx_page(struct igb_rx_buffer *rx_buffer, **/ static bool igb_add_rx_frag(struct igb_ring *rx_ring, struct igb_rx_buffer *rx_buffer, + unsigned int size, union e1000_adv_rx_desc *rx_desc, struct sk_buff *skb) { struct page *page = rx_buffer->page; unsigned char *va = page_address(page) + rx_buffer->page_offset; - unsigned int size = le16_to_cpu(rx_desc->wb.upper.length); #if (PAGE_SIZE < 8192) unsigned int truesize = IGB_RX_BUFSZ; #else @@ -6913,6 +6913,7 @@ static struct sk_buff *igb_fetch_rx_buffer(struct igb_ring *rx_ring, union e1000_adv_rx_desc *rx_desc, struct sk_buff *skb) { + unsigned int size = le16_to_cpu(rx_desc->wb.upper.length); struct igb_rx_buffer *rx_buffer; struct page *page; @@ -6948,11 +6949,11 @@ static struct sk_buff *igb_fetch_rx_buffer(struct igb_ring *rx_ring, dma_sync_single_range_for_cpu(rx_ring->dev, rx_buffer->dma, rx_buffer->page_offset, - IGB_RX_BUFSZ, + size, DMA_FROM_DEVICE); /* pull page into skb */ - if (igb_add_rx_frag(rx_ring, rx_buffer, rx_desc, skb)) { + if (igb_add_rx_frag(rx_ring, rx_buffer, size, rx_desc, skb)) { /* hand second half of page back to the ring */ igb_reuse_rx_page(rx_ring, rx_buffer); } else { -- 2.5.5