From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ben Hutchings Subject: Re: [patch net-next v2 01/15] net: introduce upper device lists Date: Tue, 14 Aug 2012 23:33:44 +0100 Message-ID: <1344983624.2690.77.camel@bwh-desktop.uk.solarflarecom.com> References: <1344956748-2099-1-git-send-email-jiri@resnulli.us> <1344956748-2099-2-git-send-email-jiri@resnulli.us> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1344956748-2099-2-git-send-email-jiri@resnulli.us> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: bridge-bounces@lists.linux-foundation.org Errors-To: bridge-bounces@lists.linux-foundation.org To: Jiri Pirko Cc: bridge@lists.linux-foundation.org, ursula.braun@de.ibm.com, john.r.fastabend@intel.com, edumazet@google.com, shemminger@vyatta.com, sean.hefty@intel.com, therbert@google.com, roland@kernel.org, linux-s390@vger.kernel.org, linux-rdma@vger.kernel.org, fubar@us.ibm.com, fbl@redhat.com, hal.rosenstock@gmail.com, faisal.latif@intel.com, linux-driver@qlogic.com, blaschka@linux.vnet.ibm.com, sony.chacko@qlogic.com, gregory.v.rose@intel.com, xiyou.wangcong@gmail.com, jitendra.kalsaria@qlogic.com, divy@chelsio.com, netdev@vger.kernel.org, linux-kernel@vger.kernel.org, kaber@trash.net, joe@perches.com, linux390@de.ibm.com, davem@davemloft.net List-Id: linux-rdma@vger.kernel.org On Tue, 2012-08-14 at 17:05 +0200, Jiri Pirko wrote: > This lists are supposed to serve for storing pointers to all upper devices. > Eventually it will replace dev->master pointer which is used for > bonding, bridge, team but it cannot be used for vlan, macvlan where > there might be multiple upper present. In case the upper link is > replacement for dev->master, it is marked with "master" flag. Something I found interesting is that the dev->master pointer and now netdev_master_upper_dev_get{,_rcu}() are hardly used by the stackled drivers that set the master. They also have to set an rx_handler on the lower device (which is itself mutually exclusive) which gets its own context pointer (rx_handler_data). Instead, the master pointer is mostly used by device drivers to find out about a bridge or bonding device above *their* devices. And that seems to work only for those specific device drivers, not e.g. openvswitch or team. I wonder if we could find a better way to encapsulate the things they want do do, in a later step (not holding up this change!). [...] > +static int __netdev_upper_dev_link(struct net_device *dev, > + struct net_device *upper_dev, bool master) > +{ > + struct netdev_upper *upper; > + > + ASSERT_RTNL(); > + > + if (dev == upper_dev) > + return -EBUSY; > + /* > + * To prevent loops, check if dev is not upper device to upper_dev. > + */ > + if (__netdev_has_upper_dev(upper_dev, dev, true)) > + return -EBUSY; [...] I think we will also need to limit the depth of the device stack so we don't run out of stack space here. __netif_receive() implements a kind of tail recursion whenever a packet is passed up, but __netdev_has_upper_dev() can't avoid doing real recursion (without the addition of a flag to net_device so it can mark its progress). Ben. -- Ben Hutchings, Staff Engineer, Solarflare Not speaking for my employer; that's the marketing department's job. They asked us to note that Solarflare product names are trademarked.