From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Thu, 20 Feb 2003 18:59:56 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Thu, 20 Feb 2003 18:59:56 -0500 Received: from palrel13.hp.com ([156.153.255.238]:51599 "EHLO palrel13.hp.com") by vger.kernel.org with ESMTP id ; Thu, 20 Feb 2003 18:59:49 -0500 Date: Thu, 20 Feb 2003 16:09:54 -0800 To: Jeff Garzik , Linux kernel mailing list Subject: [PATCH 2.5] : irtty-sir ZeroCopy Rx Message-ID: <20030221000954.GE26770@bougret.hpl.hp.com> Reply-To: jt@hpl.hp.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.3.28i Organisation: HP Labs Palo Alto Address: HP Labs, 1U-17, 1501 Page Mill road, Palo Alto, CA 94304, USA. E-mail: jt@hpl.hp.com From: Jean Tourrilhes Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org ir253_sir-dev_wrapper-4.diff : ---------------------------- o [FEATURE] Enable ZeroCopy Rx in irtty-sir/sir-dev (provided by the new SIR wrapper in 2.5.61). diff -u -p linux/drivers/net/irda/sir_dev.d7.c linux/drivers/net/irda/sir_dev.c --- linux/drivers/net/irda/sir_dev.d7.c Thu Feb 20 11:07:06 2003 +++ linux/drivers/net/irda/sir_dev.c Thu Feb 20 11:11:44 2003 @@ -223,25 +223,24 @@ int sirdev_receive(struct sir_dev *dev, } /* Read the characters into the buffer */ - while (count--) { - if (likely(atomic_read(&dev->enable_rx))) { + if (likely(atomic_read(&dev->enable_rx))) { + while (count--) /* Unwrap and destuff one byte */ async_unwrap_char(dev->netdev, &dev->stats, - &dev->rx_buff, *cp++); - } - else { + &dev->rx_buff, *cp++); + } else { + while (count--) { /* rx not enabled: save the raw bytes and never * trigger any netif_rx. The received bytes are flushed * later when we re-enable rx but might be read meanwhile * by the dongle driver. */ dev->rx_buff.data[dev->rx_buff.len++] = *cp++; - } - /* What should we do when the buffer is full? */ - if (unlikely(dev->rx_buff.len == dev->rx_buff.truesize)) - dev->rx_buff.len = 0; - + /* What should we do when the buffer is full? */ + if (unlikely(dev->rx_buff.len == dev->rx_buff.truesize)) + dev->rx_buff.len = 0; + } } return 0; @@ -423,19 +422,24 @@ static int sirdev_ioctl(struct net_devic static int sirdev_alloc_buffers(struct sir_dev *dev) { - dev->rx_buff.truesize = SIRBUF_ALLOCSIZE; dev->tx_buff.truesize = SIRBUF_ALLOCSIZE; + dev->rx_buff.truesize = IRDA_SKB_MAX_MTU; - dev->rx_buff.head = kmalloc(dev->rx_buff.truesize, GFP_KERNEL); - if (dev->rx_buff.head == NULL) + /* Bootstrap ZeroCopy Rx */ + dev->rx_buff.skb = __dev_alloc_skb(dev->rx_buff.truesize, GFP_KERNEL); + if (dev->rx_buff.skb == NULL) return -ENOMEM; - memset(dev->rx_buff.head, 0, dev->rx_buff.truesize); + skb_reserve(dev->rx_buff.skb, 1); + dev->rx_buff.head = dev->rx_buff.skb->data; + /* No need to memset the buffer, unless you are really pedantic */ dev->tx_buff.head = kmalloc(dev->tx_buff.truesize, GFP_KERNEL); if (dev->tx_buff.head == NULL) { - kfree(dev->rx_buff.head); + kfree_skb(dev->rx_buff.skb); + dev->rx_buff.skb = NULL; dev->rx_buff.head = NULL; return -ENOMEM; + /* Hu ??? This should not be here, Martin ? */ memset(dev->tx_buff.head, 0, dev->tx_buff.truesize); } @@ -451,11 +455,12 @@ static int sirdev_alloc_buffers(struct s static void sirdev_free_buffers(struct sir_dev *dev) { - if (dev->rx_buff.head) - kfree(dev->rx_buff.head); + if (dev->rx_buff.skb) + kfree_skb(dev->rx_buff.skb); if (dev->tx_buff.head) kfree(dev->tx_buff.head); dev->rx_buff.head = dev->tx_buff.head = NULL; + dev->rx_buff.skb = NULL; } static int sirdev_open(struct net_device *ndev)