From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755786AbaCCWrq (ORCPT ); Mon, 3 Mar 2014 17:47:46 -0500 Received: from mail-pd0-f179.google.com ([209.85.192.179]:47499 "EHLO mail-pd0-f179.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755755AbaCCWrm (ORCPT ); Mon, 3 Mar 2014 17:47:42 -0500 From: "Luis R. Rodriguez" To: netdev@vger.kernel.org Cc: linux-kernel@vger.kernel.org, kvm@vger.kernel.org, xen-devel@lists.xenproject.org, mcgrof@suse.com, Stephen Hemminger , bridge@lists.linux-foundation.org Subject: [RFC v3 6/6] tun: add initialization root block support Date: Mon, 3 Mar 2014 14:47:05 -0800 Message-Id: <1393886825-24323-7-git-send-email-mcgrof@do-not-panic.com> X-Mailer: git-send-email 1.8.5.3 In-Reply-To: <1393886825-24323-1-git-send-email-mcgrof@do-not-panic.com> References: <1393886825-24323-1-git-send-email-mcgrof@do-not-panic.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: "Luis R. Rodriguez" The networking bridge module allows us to specify a root block preference on net_devices but this feature is a bridge port primitive. The bridge module assumes that once a device is added as a slave to the brige that it can be considered for the the root port. Furthermore ndo_add_slave() only lets us pass on the net_device, but drivers that want to root block since the beginning need specify their root block preference prior to register_netdevice(). This means the tun driver must expose a knob to userspace to enable toggling this feature prior to adding the device to a bridge to prevent any undesired link bouncing issues on the bridge. This should be useful for TAP interfaces which are used for virtualization. Cc: Stephen Hemminger Cc: bridge@lists.linux-foundation.org Cc: netdev@vger.kernel.org Cc: linux-kernel@vger.kernel.org Cc: xen-devel@lists.xenproject.org Cc: kvm@vger.kernel.org Signed-off-by: Luis R. Rodriguez --- drivers/net/tun.c | 6 +++++- include/uapi/linux/if_tun.h | 1 + 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 8fe9cb7..e763fad 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -1693,6 +1693,9 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr) if (err < 0) goto err_free_flow; + if (ifr->ifr_flags & IFF_ROOT_BLOCK) + dev->priv_flags |= IFF_BRIDGE_ROOT_BLOCK; + err = register_netdevice(tun->dev); if (err < 0) goto err_detach; @@ -1898,7 +1901,8 @@ static long __tun_chr_ioctl(struct file *file, unsigned int cmd, * This is needed because we never checked for invalid flags on * TUNSETIFF. */ return put_user(IFF_TUN | IFF_TAP | IFF_NO_PI | IFF_ONE_QUEUE | - IFF_VNET_HDR | IFF_MULTI_QUEUE, + IFF_VNET_HDR | IFF_MULTI_QUEUE | + IFF_ROOT_BLOCK, (unsigned int __user*)argp); } else if (cmd == TUNSETQUEUE) return tun_set_queue(file, &ifr); diff --git a/include/uapi/linux/if_tun.h b/include/uapi/linux/if_tun.h index e9502dd..f814a85 100644 --- a/include/uapi/linux/if_tun.h +++ b/include/uapi/linux/if_tun.h @@ -73,6 +73,7 @@ /* read-only flag */ #define IFF_PERSIST 0x0800 #define IFF_NOFILTER 0x1000 +#define IFF_ROOT_BLOCK 0x2000 /* Socket options */ #define TUN_TX_TIMESTAMP 1 -- 1.9.0