AcoSail/kernel_optimize_test
8
0
Fork 0
forked from luck/tmp_suning_uos_patched
Code Pull Requests Activity
0456b4f8b7
kernel_optimize_test/drivers/net/tun.c

1451 lines
32 KiB
C
Raw Normal View History

Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/*
* TUN - Universal TUN/TAP device driver.
* Copyright (C) 1999-2002 Maxim Krasnyansky <maxk@qualcomm.com>
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* $Id: tun.c,v 1.15 2002/03/01 02:44:24 maxk Exp $
*/
/*
* Changes:
*
[TUNTAP]: Allow setting the linktype of the tap device from userspace Currently tun/tap only supports the EN10MB ARP type. For use with wireless and other networking types it should be possible to set the ARP type via an ioctl. Patch v2: Included check that the tap interface is down before changing the link type out from underneath it Signed-off-by: Mike Kershaw <dragorn@kismetwireless.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-09-02 08:40:05 +08:00
* Mike Kershaw <dragorn@kismetwireless.net> 2005/08/14
* Add TUNSETLINK ioctl to set the link encapsulation
*
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* Mark Smith <markzzzsmith@yahoo.com.au>
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
* Use random_ether_addr() for tap MAC address.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* Harald Roelle <harald.roelle@ifi.lmu.de> 2004/04/20
* Fixes in packet dropping, queue length setting and queue wakeup.
* Increased default tx queue length.
* Added ethtool API.
* Minor cleanups
*
* Daniel Podlejski <underley@underley.eu.org>
* Modifications for 2.3.99-pre5 kernel.
*/
#define DRV_NAME "tun"
#define DRV_VERSION "1.6"
#define DRV_DESCRIPTION "Universal TUN/TAP device driver"
#define DRV_COPYRIGHT "(C) 1999-2004 Max Krasnyansky <maxk@qualcomm.com>"
#include <linux/module.h>
#include <linux/errno.h>
#include <linux/kernel.h>
#include <linux/major.h>
#include <linux/slab.h>
net-tun: BKL pushdown Signed-off-by: Arnd Bergmann <arnd@arndb.de>
2008-05-21 01:16:24 +08:00
#include <linux/smp_lock.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
#include <linux/poll.h>
#include <linux/fcntl.h>
#include <linux/init.h>
#include <linux/skbuff.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/miscdevice.h>
#include <linux/ethtool.h>
#include <linux/rtnetlink.h>
#include <linux/if.h>
#include <linux/if_arp.h>
#include <linux/if_ether.h>
#include <linux/if_tun.h>
#include <linux/crc32.h>
[TUN]: Make the tun_dev_list per-net. Remove the static tun_dev_list and replace its occurrences in driver with per-net one. It is used in two places - in tun_set_iff and tun_cleanup. In the first case it's legal to use current net_ns. In the cleanup call - move the loop, that unregisters all devices in net exit hook. Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-16 15:41:16 +08:00
#include <linux/nsproxy.h>
tun: Allow GSO using virtio_net_hdr Add a IFF_VNET_HDR flag. This uses the same ABI as virtio_net (ie. prepending struct virtio_net_hdr to packets) to indicate GSO and checksum information. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-03 18:48:02 +08:00
#include <linux/virtio_net.h>
[NET]: Make the device list and device lookups per namespace. This patch makes most of the generic device layer network namespace safe. This patch makes dev_base_head a network namespace variable, and then it picks up a few associated variables. The functions: dev_getbyhwaddr dev_getfirsthwbytype dev_get_by_flags dev_get_by_name __dev_get_by_name dev_get_by_index __dev_get_by_index dev_ioctl dev_ethtool dev_load wireless_process_ioctl were modified to take a network namespace argument, and deal with it. vlan_ioctl_set and brioctl_set were modified so their hooks will receive a network namespace argument. So basically anthing in the core of the network stack that was affected to by the change of dev_base was modified to handle multiple network namespaces. The rest of the network stack was simply modified to explicitly use &init_net the initial network namespace. This can be fixed when those components of the network stack are modified to handle multiple network namespaces. For now the ifindex generator is left global. Fundametally ifindex numbers are per namespace, or else we will have corner case problems with migration when we get that far. At the same time there are assumptions in the network stack that the ifindex of a network device won't change. Making the ifindex number global seems a good compromise until the network stack can cope with ifindex changes when you change namespaces, and the like. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-18 02:56:21 +08:00
#include <net/net_namespace.h>
[TUN]: Introduce the tun_net structure and init/exit net ops. This is the first step in making tuntap devices work in net namespaces. The structure mentioned is pointed by generic net pointer with tun_net_id id, and tun driver fills one on its load. It will contain only the tun devices list. So declare this structure and introduce net init and exit hooks. Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-16 15:40:46 +08:00
#include <net/netns/generic.h>
tun: Implement ip link del tunXXX This greatly simplifies testing to verify I have fixed the problems with a tun device disappearing when the tun file descriptor is still held open. Further it allows removal network namespace operations for the tun driver. Reducing the network namespace handling in the driver to the minimum. i.e. When we are creating a tun device. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-22 08:02:16 +08:00
#include <net/rtnetlink.h>
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
#include <net/sock.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
#include <asm/system.h>
#include <asm/uaccess.h>
net: make struct tun_struct private to tun.c There's no reason for this to be in the header, and it just hurts recompile time. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyanskiy <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-13 09:48:58 +08:00
/* Uncomment to enable debugging */
/* #define TUN_DEBUG 1 */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
#ifdef TUN_DEBUG
static int debug;
net: make struct tun_struct private to tun.c There's no reason for this to be in the header, and it just hurts recompile time. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyanskiy <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-13 09:48:58 +08:00
#define DBG if(tun->debug)printk
#define DBG1 if(debug==2)printk
#else
#define DBG( a... )
#define DBG1( a... )
#endif
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
#define FLT_EXACT_COUNT 8
struct tap_filter {
unsigned int count; /* Number of addrs. Zero means disabled */
u32 mask[2]; /* Mask of the hashed addrs */
unsigned char addr[FLT_EXACT_COUNT][ETH_ALEN];
};
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
struct tun_file {
tun: Fix races between tun_net_close and free_netdev. The tun code does not cope gracefully if the network device goes away before the tun file descriptor is closed. It looks like we can trigger this with rmmod, and moving tun devices between network namespaces will allow this to be triggered when network namespaces exit. To fix this I introduce an intermediate data structure tun_file which holds a count of users and a pointer to the struct tun_struct. tun_get increments that reference count if it is greater than 0. tun_put decrements that reference count and detaches from the network device if the count is 0. While we have a file attached to the network device I hold a reference to the network device keeping it from going away completely. When a network device is unregistered I decrement the count of the attached tun_file and if that was the last user I detach the tun_file, and all processes on read_wait are woken up to ensure they do not sleep indefinitely. As some of those sleeps happen with the count on the tun device elevated waking up the read waiters ensures that tun_file will be detached in a timely manner. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:07:17 +08:00
atomic_t count;
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
struct tun_struct *tun;
tun: Grab the netns in open. Grabbing namespaces in open, and putting them in close always seems to be the cleanest approach with the fewest surprises. So now that we have tun_file so we have somepleace to put the network namespace, let's grab the network namespace on file open and put on file close. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:01:48 +08:00
struct net *net;
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
};
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
struct tun_sock;
net: make struct tun_struct private to tun.c There's no reason for this to be in the header, and it just hurts recompile time. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyanskiy <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-13 09:48:58 +08:00
struct tun_struct {
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
struct tun_file *tfile;
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
unsigned int flags;
net: make struct tun_struct private to tun.c There's no reason for this to be in the header, and it just hurts recompile time. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyanskiy <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-13 09:48:58 +08:00
uid_t owner;
gid_t group;
struct sk_buff_head readq;
struct net_device *dev;
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
struct fasync_struct *fasync;
net: make struct tun_struct private to tun.c There's no reason for this to be in the header, and it just hurts recompile time. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyanskiy <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-13 09:48:58 +08:00
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
struct tap_filter txflt;
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
struct sock *sk;
struct socket socket;
net: make struct tun_struct private to tun.c There's no reason for this to be in the header, and it just hurts recompile time. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyanskiy <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-13 09:48:58 +08:00
#ifdef TUN_DEBUG
int debug;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
#endif
net: make struct tun_struct private to tun.c There's no reason for this to be in the header, and it just hurts recompile time. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyanskiy <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-13 09:48:58 +08:00
};
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
struct tun_sock {
struct sock sk;
struct tun_struct *tun;
};
static inline struct tun_sock *tun_sk(struct sock *sk)
{
return container_of(sk, struct tun_sock, sk);
}
tun: Fix races in tun_set_iff It is possible for two different tasks with access to the same file descriptor to call tun_set_iff on it at the same time and race to attach to a tap device. Prevent this by placing all of the logic to attach to a file descriptor in one function and testing the file descriptor to be certain it is not already attached to another tun device. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 18:57:48 +08:00
static int tun_attach(struct tun_struct *tun, struct file *file)
{
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
struct tun_file *tfile = file->private_data;
tun: Fix races in tun_set_iff It is possible for two different tasks with access to the same file descriptor to call tun_set_iff on it at the same time and race to attach to a tap device. Prevent this by placing all of the logic to attach to a file descriptor in one function and testing the file descriptor to be certain it is not already attached to another tun device. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 18:57:48 +08:00
const struct cred *cred = current_cred();
tun: Make tun_net_xmit atomic wrt tun_attach && tun_detach Currently this small race allows for a packet to be received when we detach from an tun device and still be enqueued. Not especially important but not what the code is trying to do. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:02:28 +08:00
int err;
tun: Fix races in tun_set_iff It is possible for two different tasks with access to the same file descriptor to call tun_set_iff on it at the same time and race to attach to a tap device. Prevent this by placing all of the logic to attach to a file descriptor in one function and testing the file descriptor to be certain it is not already attached to another tun device. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 18:57:48 +08:00
ASSERT_RTNL();
/* Check permissions */
if (((tun->owner != -1 && cred->euid != tun->owner) ||
tun: Check supplemental groups in TUN/TAP driver. Michael Tokarev wrote: [] > 2, and this is the main one: How about supplementary groups? > > Here I have a valid usage case: a group of testers running various > versions of windows using KVM (kernel virtual machine), 1 at a time, > to test some software. kvm is set up to use bridge with a tap device > (there should be a way to connect to the machine). Anyone on that group > has to be able to start/stop the virtual machines. > > My first attempt - pretty obvious when I saw -g option of tunctl - is > to add group ownership for the tun device and add a supplementary group > to each user (their primary group should be different). But that fails, > since kernel only checks for egid, not any other group ids. > > What's the reasoning to not allow supplementary groups and to only check > for egid? Signed-off-by: Michael Tokarev <mjt@tls.msk.ru> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-03 15:34:56 +08:00
(tun->group != -1 && !in_egroup_p(tun->group))) &&
tun: Fix races in tun_set_iff It is possible for two different tasks with access to the same file descriptor to call tun_set_iff on it at the same time and race to attach to a tap device. Prevent this by placing all of the logic to attach to a file descriptor in one function and testing the file descriptor to be certain it is not already attached to another tun device. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 18:57:48 +08:00
!capable(CAP_NET_ADMIN))
return -EPERM;
tun: Make tun_net_xmit atomic wrt tun_attach && tun_detach Currently this small race allows for a packet to be received when we detach from an tun device and still be enqueued. Not especially important but not what the code is trying to do. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:02:28 +08:00
netif_tx_lock_bh(tun->dev);
err = -EINVAL;
if (tfile->tun)
goto out;
err = -EBUSY;
if (tun->tfile)
goto out;
err = 0;
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
tfile->tun = tun;
tun->tfile = tfile;
tun: Fix races between tun_net_close and free_netdev. The tun code does not cope gracefully if the network device goes away before the tun file descriptor is closed. It looks like we can trigger this with rmmod, and moving tun devices between network namespaces will allow this to be triggered when network namespaces exit. To fix this I introduce an intermediate data structure tun_file which holds a count of users and a pointer to the struct tun_struct. tun_get increments that reference count if it is greater than 0. tun_put decrements that reference count and detaches from the network device if the count is 0. While we have a file attached to the network device I hold a reference to the network device keeping it from going away completely. When a network device is unregistered I decrement the count of the attached tun_file and if that was the last user I detach the tun_file, and all processes on read_wait are woken up to ensure they do not sleep indefinitely. As some of those sleeps happen with the count on the tun device elevated waking up the read waiters ensures that tun_file will be detached in a timely manner. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:07:17 +08:00
dev_hold(tun->dev);
tun: Only free a netdev when all tun descriptors are closed The commit c70f182940f988448f3c12a209d18b1edc276e33 ("tun: Fix races between tun_net_close and free_netdev") fixed a race where an asynchronous deletion of a tun device can hose a poll(2) on a tun fd attached to that device. However, this came at the cost of moving the tun wait queue into the tun file data structure. The problem with this is that it imposes restrictions on when and where the tun device can access the wait queue since the tun file may change at any time due to detaching and reattaching. In particular, now that we need to use the wait queue on the receive path it becomes difficult to properly synchronise this with the detachment of the tun device. This patch solves the original race in a different way. Since the race is only because the underlying memory gets freed, we can prevent it simply by ensuring that we don't do that until all tun descriptors ever attached to the device (even if they have since be detached because they may still be sitting in poll) have been closed. This is done by using reference counting the attached tun file descriptors. The refcount in tun->sk has been reappropriated for this purpose since it was already being used for that, albeit from the opposite angle. Note that we no longer zero tfile->tun since tun_get will return NULL anyway after the refcount on tfile hits zero. Instead it represents whether this device has ever been attached to a device. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-18 22:15:52 +08:00
sock_hold(tun->sk);
tun: Fix races between tun_net_close and free_netdev. The tun code does not cope gracefully if the network device goes away before the tun file descriptor is closed. It looks like we can trigger this with rmmod, and moving tun devices between network namespaces will allow this to be triggered when network namespaces exit. To fix this I introduce an intermediate data structure tun_file which holds a count of users and a pointer to the struct tun_struct. tun_get increments that reference count if it is greater than 0. tun_put decrements that reference count and detaches from the network device if the count is 0. While we have a file attached to the network device I hold a reference to the network device keeping it from going away completely. When a network device is unregistered I decrement the count of the attached tun_file and if that was the last user I detach the tun_file, and all processes on read_wait are woken up to ensure they do not sleep indefinitely. As some of those sleeps happen with the count on the tun device elevated waking up the read waiters ensures that tun_file will be detached in a timely manner. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:07:17 +08:00
atomic_inc(&tfile->count);
tun: Fix races in tun_set_iff It is possible for two different tasks with access to the same file descriptor to call tun_set_iff on it at the same time and race to attach to a tap device. Prevent this by placing all of the logic to attach to a file descriptor in one function and testing the file descriptor to be certain it is not already attached to another tun device. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 18:57:48 +08:00
tun: Make tun_net_xmit atomic wrt tun_attach && tun_detach Currently this small race allows for a packet to be received when we detach from an tun device and still be enqueued. Not especially important but not what the code is trying to do. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:02:28 +08:00
out:
netif_tx_unlock_bh(tun->dev);
return err;
tun: Fix races in tun_set_iff It is possible for two different tasks with access to the same file descriptor to call tun_set_iff on it at the same time and race to attach to a tap device. Prevent this by placing all of the logic to attach to a file descriptor in one function and testing the file descriptor to be certain it is not already attached to another tun device. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 18:57:48 +08:00
}
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
static void __tun_detach(struct tun_struct *tun)
{
/* Detach from net device */
tun: Make tun_net_xmit atomic wrt tun_attach && tun_detach Currently this small race allows for a packet to be received when we detach from an tun device and still be enqueued. Not especially important but not what the code is trying to do. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:02:28 +08:00
netif_tx_lock_bh(tun->dev);
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
tun->tfile = NULL;
tun: Make tun_net_xmit atomic wrt tun_attach && tun_detach Currently this small race allows for a packet to be received when we detach from an tun device and still be enqueued. Not especially important but not what the code is trying to do. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:02:28 +08:00
netif_tx_unlock_bh(tun->dev);
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
/* Drop read queue */
skb_queue_purge(&tun->readq);
tun: Fix races between tun_net_close and free_netdev. The tun code does not cope gracefully if the network device goes away before the tun file descriptor is closed. It looks like we can trigger this with rmmod, and moving tun devices between network namespaces will allow this to be triggered when network namespaces exit. To fix this I introduce an intermediate data structure tun_file which holds a count of users and a pointer to the struct tun_struct. tun_get increments that reference count if it is greater than 0. tun_put decrements that reference count and detaches from the network device if the count is 0. While we have a file attached to the network device I hold a reference to the network device keeping it from going away completely. When a network device is unregistered I decrement the count of the attached tun_file and if that was the last user I detach the tun_file, and all processes on read_wait are woken up to ensure they do not sleep indefinitely. As some of those sleeps happen with the count on the tun device elevated waking up the read waiters ensures that tun_file will be detached in a timely manner. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:07:17 +08:00
/* Drop the extra count on the net device */
dev_put(tun->dev);
}
static void tun_detach(struct tun_struct *tun)
{
rtnl_lock();
__tun_detach(tun);
rtnl_unlock();
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
}
static struct tun_struct *__tun_get(struct tun_file *tfile)
{
tun: Fix races between tun_net_close and free_netdev. The tun code does not cope gracefully if the network device goes away before the tun file descriptor is closed. It looks like we can trigger this with rmmod, and moving tun devices between network namespaces will allow this to be triggered when network namespaces exit. To fix this I introduce an intermediate data structure tun_file which holds a count of users and a pointer to the struct tun_struct. tun_get increments that reference count if it is greater than 0. tun_put decrements that reference count and detaches from the network device if the count is 0. While we have a file attached to the network device I hold a reference to the network device keeping it from going away completely. When a network device is unregistered I decrement the count of the attached tun_file and if that was the last user I detach the tun_file, and all processes on read_wait are woken up to ensure they do not sleep indefinitely. As some of those sleeps happen with the count on the tun device elevated waking up the read waiters ensures that tun_file will be detached in a timely manner. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:07:17 +08:00
struct tun_struct *tun = NULL;
if (atomic_inc_not_zero(&tfile->count))
tun = tfile->tun;
return tun;
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
}
static struct tun_struct *tun_get(struct file *file)
{
return __tun_get(file->private_data);
}
static void tun_put(struct tun_struct *tun)
{
tun: Fix races between tun_net_close and free_netdev. The tun code does not cope gracefully if the network device goes away before the tun file descriptor is closed. It looks like we can trigger this with rmmod, and moving tun devices between network namespaces will allow this to be triggered when network namespaces exit. To fix this I introduce an intermediate data structure tun_file which holds a count of users and a pointer to the struct tun_struct. tun_get increments that reference count if it is greater than 0. tun_put decrements that reference count and detaches from the network device if the count is 0. While we have a file attached to the network device I hold a reference to the network device keeping it from going away completely. When a network device is unregistered I decrement the count of the attached tun_file and if that was the last user I detach the tun_file, and all processes on read_wait are woken up to ensure they do not sleep indefinitely. As some of those sleeps happen with the count on the tun device elevated waking up the read waiters ensures that tun_file will be detached in a timely manner. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:07:17 +08:00
struct tun_file *tfile = tun->tfile;
if (atomic_dec_and_test(&tfile->count))
tun_detach(tfile->tun);
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
}
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
/* TAP filterting */
static void addr_hash_set(u32 *mask, const u8 *addr)
{
int n = ether_crc(ETH_ALEN, addr) >> 26;
mask[n >> 5] |= (1 << (n & 31));
}
static unsigned int addr_hash_test(const u32 *mask, const u8 *addr)
{
int n = ether_crc(ETH_ALEN, addr) >> 26;
return mask[n >> 5] & (1 << (n & 31));
}
static int update_filter(struct tap_filter *filter, void __user *arg)
{
struct { u8 u[ETH_ALEN]; } *addr;
struct tun_filter uf;
int err, alen, n, nexact;
if (copy_from_user(&uf, arg, sizeof(uf)))
return -EFAULT;
if (!uf.count) {
/* Disabled */
filter->count = 0;
return 0;
}
alen = ETH_ALEN * uf.count;
addr = kmalloc(alen, GFP_KERNEL);
if (!addr)
return -ENOMEM;
if (copy_from_user(addr, arg + sizeof(uf), alen)) {
err = -EFAULT;
goto done;
}
/* The filter is updated without holding any locks. Which is
* perfectly safe. We disable it first and in the worst
* case we'll accept a few undesired packets. */
filter->count = 0;
wmb();
/* Use first set of addresses as an exact filter */
for (n = 0; n < uf.count && n < FLT_EXACT_COUNT; n++)
memcpy(filter->addr[n], addr[n].u, ETH_ALEN);
nexact = n;
tun: Fix unicast filter overflow Tap devices can make use of a small MAC filter set via the TUNSETTXFILTER ioctl. The filter has a set of exact matches plus a hash for imperfect filtering of additional multicast addresses. The current code is unbalanced, adding unicast addresses to the multicast hash, but only checking the hash against multicast addresses. This results in the filter dropping unicast addresses that overflow the exact filter. The fix is simply to disable the filter by leaving count set to zero if we find non-multicast addresses after the exact match table is filled. Signed-off-by: Alex Williamson <alex.williamson@hp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-09 09:49:17 +08:00
/* Remaining multicast addresses are hashed,
* unicast will leave the filter disabled. */
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
memset(filter->mask, 0, sizeof(filter->mask));
tun: Fix unicast filter overflow Tap devices can make use of a small MAC filter set via the TUNSETTXFILTER ioctl. The filter has a set of exact matches plus a hash for imperfect filtering of additional multicast addresses. The current code is unbalanced, adding unicast addresses to the multicast hash, but only checking the hash against multicast addresses. This results in the filter dropping unicast addresses that overflow the exact filter. The fix is simply to disable the filter by leaving count set to zero if we find non-multicast addresses after the exact match table is filled. Signed-off-by: Alex Williamson <alex.williamson@hp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-09 09:49:17 +08:00
for (; n < uf.count; n++) {
if (!is_multicast_ether_addr(addr[n].u)) {
err = 0; /* no filter */
goto done;
}
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
addr_hash_set(filter->mask, addr[n].u);
tun: Fix unicast filter overflow Tap devices can make use of a small MAC filter set via the TUNSETTXFILTER ioctl. The filter has a set of exact matches plus a hash for imperfect filtering of additional multicast addresses. The current code is unbalanced, adding unicast addresses to the multicast hash, but only checking the hash against multicast addresses. This results in the filter dropping unicast addresses that overflow the exact filter. The fix is simply to disable the filter by leaving count set to zero if we find non-multicast addresses after the exact match table is filled. Signed-off-by: Alex Williamson <alex.williamson@hp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-09 09:49:17 +08:00
}
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
/* For ALLMULTI just set the mask to all ones.
* This overrides the mask populated above. */
if ((uf.flags & TUN_FLT_ALLMULTI))
memset(filter->mask, ~0, sizeof(filter->mask));
/* Now enable the filter */
wmb();
filter->count = nexact;
/* Return the number of exact filters */
err = nexact;
done:
kfree(addr);
return err;
}
/* Returns: 0 - drop, !=0 - accept */
static int run_filter(struct tap_filter *filter, const struct sk_buff *skb)
{
/* Cannot use eth_hdr(skb) here because skb_mac_hdr() is incorrect
* at this point. */
struct ethhdr *eh = (struct ethhdr *) skb->data;
int i;
/* Exact match */
for (i = 0; i < filter->count; i++)
if (!compare_ether_addr(eh->h_dest, filter->addr[i]))
return 1;
/* Inexact match (multicast only) */
if (is_multicast_ether_addr(eh->h_dest))
return addr_hash_test(filter->mask, eh->h_dest);
return 0;
}
/*
* Checks whether the packet is accepted or not.
* Returns: 0 - drop, !=0 - accept
*/
static int check_filter(struct tap_filter *filter, const struct sk_buff *skb)
{
if (!filter->count)
return 1;
return run_filter(filter, skb);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* Network device part of the driver */
drivers/net: const-ify ethtool_ops declarations Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 02:30:00 +08:00
static const struct ethtool_ops tun_ethtool_ops;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
tun: Fix races between tun_net_close and free_netdev. The tun code does not cope gracefully if the network device goes away before the tun file descriptor is closed. It looks like we can trigger this with rmmod, and moving tun devices between network namespaces will allow this to be triggered when network namespaces exit. To fix this I introduce an intermediate data structure tun_file which holds a count of users and a pointer to the struct tun_struct. tun_get increments that reference count if it is greater than 0. tun_put decrements that reference count and detaches from the network device if the count is 0. While we have a file attached to the network device I hold a reference to the network device keeping it from going away completely. When a network device is unregistered I decrement the count of the attached tun_file and if that was the last user I detach the tun_file, and all processes on read_wait are woken up to ensure they do not sleep indefinitely. As some of those sleeps happen with the count on the tun device elevated waking up the read waiters ensures that tun_file will be detached in a timely manner. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:07:17 +08:00
/* Net device detach from fd. */
static void tun_net_uninit(struct net_device *dev)
{
struct tun_struct *tun = netdev_priv(dev);
struct tun_file *tfile = tun->tfile;
/* Inform the methods they need to stop using the dev.
*/
if (tfile) {
tun: Fix sk_sleep races when attaching/detaching As the sk_sleep wait queue actually lives in tfile, which may be detached from the tun device, bad things will happen when we use sk_sleep after detaching. Since the tun device is the persistent data structure here (when requested by the user), it makes much more sense to have the wait queue live there. There is no reason to have it in tfile at all since the only time we can wait is if we have a tun attached. In fact we already have a wait queue in tun_struct, so we might as well use it. Reported-by: Eric W. Biederman <ebiederm@xmission.com> Tested-by: Christian Borntraeger <borntraeger@de.ibm.com> Tested-by: Patrick McHardy <kaber@trash.net> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 06:35:50 +08:00
wake_up_all(&tun->socket.wait);
tun: Fix races between tun_net_close and free_netdev. The tun code does not cope gracefully if the network device goes away before the tun file descriptor is closed. It looks like we can trigger this with rmmod, and moving tun devices between network namespaces will allow this to be triggered when network namespaces exit. To fix this I introduce an intermediate data structure tun_file which holds a count of users and a pointer to the struct tun_struct. tun_get increments that reference count if it is greater than 0. tun_put decrements that reference count and detaches from the network device if the count is 0. While we have a file attached to the network device I hold a reference to the network device keeping it from going away completely. When a network device is unregistered I decrement the count of the attached tun_file and if that was the last user I detach the tun_file, and all processes on read_wait are woken up to ensure they do not sleep indefinitely. As some of those sleeps happen with the count on the tun device elevated waking up the read waiters ensures that tun_file will be detached in a timely manner. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:07:17 +08:00
if (atomic_dec_and_test(&tfile->count))
__tun_detach(tun);
}
}
tun: Only free a netdev when all tun descriptors are closed The commit c70f182940f988448f3c12a209d18b1edc276e33 ("tun: Fix races between tun_net_close and free_netdev") fixed a race where an asynchronous deletion of a tun device can hose a poll(2) on a tun fd attached to that device. However, this came at the cost of moving the tun wait queue into the tun file data structure. The problem with this is that it imposes restrictions on when and where the tun device can access the wait queue since the tun file may change at any time due to detaching and reattaching. In particular, now that we need to use the wait queue on the receive path it becomes difficult to properly synchronise this with the detachment of the tun device. This patch solves the original race in a different way. Since the race is only because the underlying memory gets freed, we can prevent it simply by ensuring that we don't do that until all tun descriptors ever attached to the device (even if they have since be detached because they may still be sitting in poll) have been closed. This is done by using reference counting the attached tun file descriptors. The refcount in tun->sk has been reappropriated for this purpose since it was already being used for that, albeit from the opposite angle. Note that we no longer zero tfile->tun since tun_get will return NULL anyway after the refcount on tfile hits zero. Instead it represents whether this device has ever been attached to a device. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-18 22:15:52 +08:00
static void tun_free_netdev(struct net_device *dev)
{
struct tun_struct *tun = netdev_priv(dev);
sock_put(tun->sk);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* Net device open. */
static int tun_net_open(struct net_device *dev)
{
netif_start_queue(dev);
return 0;
}
/* Net device close. */
static int tun_net_close(struct net_device *dev)
{
netif_stop_queue(dev);
return 0;
}
/* Net device start xmit */
static int tun_net_xmit(struct sk_buff *skb, struct net_device *dev)
{
struct tun_struct *tun = netdev_priv(dev);
DBG(KERN_INFO "%s: tun_net_xmit %d\n", tun->dev->name, skb->len);
/* Drop packet if interface is not attached */
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
if (!tun->tfile)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
goto drop;
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
/* Drop if the filter does not like it.
* This is a noop if the filter is disabled.
* Filter can be enabled only for the TAP devices. */
if (!check_filter(&tun->txflt, skb))
goto drop;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (skb_queue_len(&tun->readq) >= dev->tx_queue_len) {
if (!(tun->flags & TUN_ONE_QUEUE)) {
/* Normal queueing mode. */
/* Packet scheduler handles dropping of further packets. */
netif_stop_queue(dev);
/* We won't see all dropped packets individually, so overrun
* error is more appropriate. */
[NET] drivers/net: statistics cleanup #1 -- save memory and shrink code We now have struct net_device_stats embedded in struct net_device, and the default ->get_stats() hook does the obvious thing for us. Run through drivers/net/* and remove the driver-local storage of statistics, and driver-local ->get_stats() hook where applicable. This was just the low-hanging fruit in drivers/net; plenty more drivers remain to be updated. [ Resolved conflicts with napi_struct changes and fix sunqe build regression... -DaveM ] Signed-off-by: Jeff Garzik <jeff@garzik.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 08:41:50 +08:00
dev->stats.tx_fifo_errors++;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
} else {
/* Single queue mode.
* Driver handles dropping of all packets itself. */
goto drop;
}
}
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
/* Enqueue packet */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
skb_queue_tail(&tun->readq, skb);
dev->trans_start = jiffies;
/* Notify and wake up reader process */
if (tun->flags & TUN_FASYNC)
kill_fasync(&tun->fasync, SIGIO, POLL_IN);
tun: Fix sk_sleep races when attaching/detaching As the sk_sleep wait queue actually lives in tfile, which may be detached from the tun device, bad things will happen when we use sk_sleep after detaching. Since the tun device is the persistent data structure here (when requested by the user), it makes much more sense to have the wait queue live there. There is no reason to have it in tfile at all since the only time we can wait is if we have a tun attached. In fact we already have a wait queue in tun_struct, so we might as well use it. Reported-by: Eric W. Biederman <ebiederm@xmission.com> Tested-by: Christian Borntraeger <borntraeger@de.ibm.com> Tested-by: Patrick McHardy <kaber@trash.net> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 06:35:50 +08:00
wake_up_interruptible(&tun->socket.wait);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return 0;
drop:
[NET] drivers/net: statistics cleanup #1 -- save memory and shrink code We now have struct net_device_stats embedded in struct net_device, and the default ->get_stats() hook does the obvious thing for us. Run through drivers/net/* and remove the driver-local storage of statistics, and driver-local ->get_stats() hook where applicable. This was just the low-hanging fruit in drivers/net; plenty more drivers remain to be updated. [ Resolved conflicts with napi_struct changes and fix sunqe build regression... -DaveM ] Signed-off-by: Jeff Garzik <jeff@garzik.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 08:41:50 +08:00
dev->stats.tx_dropped++;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
kfree_skb(skb);
return 0;
}
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
static void tun_net_mclist(struct net_device *dev)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
/*
* This callback is supposed to deal with mc filter in
* _rx_ path and has nothing to do with the _tx_ path.
* In rx path we always accept everything userspace gives us.
*/
return;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
[TAP]: Configurable interface MTU. Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-17 03:21:38 +08:00
#define MIN_MTU 68
#define MAX_MTU 65535
static int
tun_net_change_mtu(struct net_device *dev, int new_mtu)
{
if (new_mtu < MIN_MTU || new_mtu + dev->hard_header_len > MAX_MTU)
return -EINVAL;
dev->mtu = new_mtu;
return 0;
}
tun: convert to net_device_ops Convert the TUN/TAP tunnel driver to net_device_ops. Split the ops in two, and retain compatability. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 14:10:37 +08:00
static const struct net_device_ops tun_netdev_ops = {
tun: Fix races between tun_net_close and free_netdev. The tun code does not cope gracefully if the network device goes away before the tun file descriptor is closed. It looks like we can trigger this with rmmod, and moving tun devices between network namespaces will allow this to be triggered when network namespaces exit. To fix this I introduce an intermediate data structure tun_file which holds a count of users and a pointer to the struct tun_struct. tun_get increments that reference count if it is greater than 0. tun_put decrements that reference count and detaches from the network device if the count is 0. While we have a file attached to the network device I hold a reference to the network device keeping it from going away completely. When a network device is unregistered I decrement the count of the attached tun_file and if that was the last user I detach the tun_file, and all processes on read_wait are woken up to ensure they do not sleep indefinitely. As some of those sleeps happen with the count on the tun device elevated waking up the read waiters ensures that tun_file will be detached in a timely manner. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:07:17 +08:00
.ndo_uninit = tun_net_uninit,
tun: convert to net_device_ops Convert the TUN/TAP tunnel driver to net_device_ops. Split the ops in two, and retain compatability. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 14:10:37 +08:00
.ndo_open = tun_net_open,
.ndo_stop = tun_net_close,
netdev: add more functions to netdevice ops This patch moves neigh_setup and hard_start_xmit into the network device ops structure. For bisection, fix all the previously converted drivers as well. Bonding driver took the biggest hit on this. Added a prefetch of the hard_start_xmit in the fast path to try and reduce any impact this would have. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-21 12:14:53 +08:00
.ndo_start_xmit = tun_net_xmit,
tun: convert to net_device_ops Convert the TUN/TAP tunnel driver to net_device_ops. Split the ops in two, and retain compatability. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 14:10:37 +08:00
.ndo_change_mtu = tun_net_change_mtu,
};
static const struct net_device_ops tap_netdev_ops = {
tun: Fix races between tun_net_close and free_netdev. The tun code does not cope gracefully if the network device goes away before the tun file descriptor is closed. It looks like we can trigger this with rmmod, and moving tun devices between network namespaces will allow this to be triggered when network namespaces exit. To fix this I introduce an intermediate data structure tun_file which holds a count of users and a pointer to the struct tun_struct. tun_get increments that reference count if it is greater than 0. tun_put decrements that reference count and detaches from the network device if the count is 0. While we have a file attached to the network device I hold a reference to the network device keeping it from going away completely. When a network device is unregistered I decrement the count of the attached tun_file and if that was the last user I detach the tun_file, and all processes on read_wait are woken up to ensure they do not sleep indefinitely. As some of those sleeps happen with the count on the tun device elevated waking up the read waiters ensures that tun_file will be detached in a timely manner. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:07:17 +08:00
.ndo_uninit = tun_net_uninit,
tun: convert to net_device_ops Convert the TUN/TAP tunnel driver to net_device_ops. Split the ops in two, and retain compatability. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 14:10:37 +08:00
.ndo_open = tun_net_open,
.ndo_stop = tun_net_close,
netdev: add more functions to netdevice ops This patch moves neigh_setup and hard_start_xmit into the network device ops structure. For bisection, fix all the previously converted drivers as well. Bonding driver took the biggest hit on this. Added a prefetch of the hard_start_xmit in the fast path to try and reduce any impact this would have. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-21 12:14:53 +08:00
.ndo_start_xmit = tun_net_xmit,
tun: convert to net_device_ops Convert the TUN/TAP tunnel driver to net_device_ops. Split the ops in two, and retain compatability. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 14:10:37 +08:00
.ndo_change_mtu = tun_net_change_mtu,
.ndo_set_multicast_list = tun_net_mclist,
.ndo_set_mac_address = eth_mac_addr,
.ndo_validate_addr = eth_validate_addr,
};
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* Initialize net device. */
static void tun_net_init(struct net_device *dev)
{
struct tun_struct *tun = netdev_priv(dev);
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
switch (tun->flags & TUN_TYPE_MASK) {
case TUN_TUN_DEV:
tun: convert to net_device_ops Convert the TUN/TAP tunnel driver to net_device_ops. Split the ops in two, and retain compatability. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 14:10:37 +08:00
dev->netdev_ops = &tun_netdev_ops;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* Point-to-Point TUN Device */
dev->hard_header_len = 0;
dev->addr_len = 0;
dev->mtu = 1500;
/* Zero header length */
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
dev->type = ARPHRD_NONE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
dev->flags = IFF_POINTOPOINT | IFF_NOARP | IFF_MULTICAST;
dev->tx_queue_len = TUN_READQ_SIZE; /* We prefer our own queue length */
break;
case TUN_TAP_DEV:
tun: Fix SIOCSIFHWADDR error. Set proper operations. Signed-off-by: Kusanagi Kouichi <slash@ma.neweb.ne.jp> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-30 10:23:28 +08:00
dev->netdev_ops = &tap_netdev_ops;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* Ethernet TAP Device */
ether_setup(dev);
[NET] tun/tap: fixed hw address handling Fixed tun/tap driver's handling of hw addresses. The hw address is stored in both the net_device.dev_addr and tun.dev_addr fields. These fields were not kept synchronized, and in fact weren't even initialized to the same value. Now during both init and when performing SIOCSIFHWADDR on the tun device these values are both updated. However, if SIOCSIFHWADDR is performed on the net device directly (for instance, setting the hw address using ifconfig), the tun device does not get updated. Perhaps the tun.dev_addr field should be removed completely at some point, as it is redundant and net_device.dev_addr can be used anywhere it is used. Signed-off-by: Brian Braunstein <linuxkernel@bristyle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-04-26 16:00:55 +08:00
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
random_ether_addr(dev->dev_addr);
[NET] tun/tap: fixed hw address handling Fixed tun/tap driver's handling of hw addresses. The hw address is stored in both the net_device.dev_addr and tun.dev_addr fields. These fields were not kept synchronized, and in fact weren't even initialized to the same value. Now during both init and when performing SIOCSIFHWADDR on the tun device these values are both updated. However, if SIOCSIFHWADDR is performed on the net device directly (for instance, setting the hw address using ifconfig), the tun device does not get updated. Perhaps the tun.dev_addr field should be removed completely at some point, as it is redundant and net_device.dev_addr can be used anywhere it is used. Signed-off-by: Brian Braunstein <linuxkernel@bristyle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-04-26 16:00:55 +08:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
dev->tx_queue_len = TUN_READQ_SIZE; /* We prefer our own queue length */
break;
}
}
/* Character device part */
/* Poll */
static unsigned int tun_chr_poll(struct file *file, poll_table * wait)
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
{
tun: Move read_wait into tun_file The poll interface requires that the waitqueue exist while the struct file is open. In the rare case when a tun device disappears before the tun file closes we fail to provide this property, so move read_wait. This is safe now that tun_net_xmit is atomic with tun_detach. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:03:21 +08:00
struct tun_file *tfile = file->private_data;
struct tun_struct *tun = __tun_get(tfile);
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
struct sock *sk = tun->sk;
unsigned int mask = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (!tun)
tun: Use POLLERR not EBADF in tun_chr_poll EBADF is meaningless in the context of a poll mask so use POLLERR instead. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 18:59:05 +08:00
return POLLERR;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
DBG(KERN_INFO "%s: tun_chr_poll\n", tun->dev->name);
tun: Fix sk_sleep races when attaching/detaching As the sk_sleep wait queue actually lives in tfile, which may be detached from the tun device, bad things will happen when we use sk_sleep after detaching. Since the tun device is the persistent data structure here (when requested by the user), it makes much more sense to have the wait queue live there. There is no reason to have it in tfile at all since the only time we can wait is if we have a tun attached. In fact we already have a wait queue in tun_struct, so we might as well use it. Reported-by: Eric W. Biederman <ebiederm@xmission.com> Tested-by: Christian Borntraeger <borntraeger@de.ibm.com> Tested-by: Patrick McHardy <kaber@trash.net> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 06:35:50 +08:00
poll_wait(file, &tun->socket.wait, wait);
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
[NET]: Transform skb_queue_len() binary tests into skb_queue_empty() This is part of the grand scheme to eliminate the qlen member of skb_queue_head, and subsequently remove the 'list' member of sk_buff. Most users of skb_queue_len() want to know if the queue is empty or not, and that's trivially done with skb_queue_empty() which doesn't use the skb_queue_head->qlen member and instead uses the queue list emptyness as the test. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-07-09 05:57:23 +08:00
if (!skb_queue_empty(&tun->readq))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
mask |= POLLIN | POLLRDNORM;
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
if (sock_writeable(sk) ||
(!test_and_set_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags) &&
sock_writeable(sk)))
mask |= POLLOUT | POLLWRNORM;
tun: Fix races between tun_net_close and free_netdev. The tun code does not cope gracefully if the network device goes away before the tun file descriptor is closed. It looks like we can trigger this with rmmod, and moving tun devices between network namespaces will allow this to be triggered when network namespaces exit. To fix this I introduce an intermediate data structure tun_file which holds a count of users and a pointer to the struct tun_struct. tun_get increments that reference count if it is greater than 0. tun_put decrements that reference count and detaches from the network device if the count is 0. While we have a file attached to the network device I hold a reference to the network device keeping it from going away completely. When a network device is unregistered I decrement the count of the attached tun_file and if that was the last user I detach the tun_file, and all processes on read_wait are woken up to ensure they do not sleep indefinitely. As some of those sleeps happen with the count on the tun device elevated waking up the read waiters ensures that tun_file will be detached in a timely manner. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:07:17 +08:00
if (tun->dev->reg_state != NETREG_REGISTERED)
mask = POLLERR;
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
tun_put(tun);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return mask;
}
tun: fallback if skb_alloc() fails on big packets skb_alloc produces linear packets (using kmalloc()). That can fail, so should we fall back to making paged skbs. My original version of this patch always allocate paged skbs for big packets. But that made performance drop from 8.4 seconds to 8.8 seconds on 1G lguest->Host TCP xmit. So now we only do that as a fallback. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-16 06:15:10 +08:00
/* prepad is the amount to reserve at front. len is length after that.
* linear is a hint as to how much to copy (usually headers). */
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
static inline struct sk_buff *tun_alloc_skb(struct tun_struct *tun,
size_t prepad, size_t len,
size_t linear, int noblock)
tun: fallback if skb_alloc() fails on big packets skb_alloc produces linear packets (using kmalloc()). That can fail, so should we fall back to making paged skbs. My original version of this patch always allocate paged skbs for big packets. But that made performance drop from 8.4 seconds to 8.8 seconds on 1G lguest->Host TCP xmit. So now we only do that as a fallback. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-16 06:15:10 +08:00
{
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
struct sock *sk = tun->sk;
tun: fallback if skb_alloc() fails on big packets skb_alloc produces linear packets (using kmalloc()). That can fail, so should we fall back to making paged skbs. My original version of this patch always allocate paged skbs for big packets. But that made performance drop from 8.4 seconds to 8.8 seconds on 1G lguest->Host TCP xmit. So now we only do that as a fallback. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-16 06:15:10 +08:00
struct sk_buff *skb;
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
int err;
tun: fallback if skb_alloc() fails on big packets skb_alloc produces linear packets (using kmalloc()). That can fail, so should we fall back to making paged skbs. My original version of this patch always allocate paged skbs for big packets. But that made performance drop from 8.4 seconds to 8.8 seconds on 1G lguest->Host TCP xmit. So now we only do that as a fallback. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-16 06:15:10 +08:00
/* Under a page? Don't bother with paged skb. */
tun: Fix crash with non-GSO users When I made the tun driver use non-linear packets as the preferred option, it broke non-GSO users because they would end up allocating a completely non-linear packet, which triggers a crash when we call eth_type_trans. This patch reverts non-GSO users to using linear packets and adds a check to ensure that GSO users can't cause crashes in the same way. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-14 17:09:43 +08:00
if (prepad + len < PAGE_SIZE || !linear)
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
linear = len;
tun: fallback if skb_alloc() fails on big packets skb_alloc produces linear packets (using kmalloc()). That can fail, so should we fall back to making paged skbs. My original version of this patch always allocate paged skbs for big packets. But that made performance drop from 8.4 seconds to 8.8 seconds on 1G lguest->Host TCP xmit. So now we only do that as a fallback. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-16 06:15:10 +08:00
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
skb = sock_alloc_send_pskb(sk, prepad + linear, len - linear, noblock,
&err);
tun: fallback if skb_alloc() fails on big packets skb_alloc produces linear packets (using kmalloc()). That can fail, so should we fall back to making paged skbs. My original version of this patch always allocate paged skbs for big packets. But that made performance drop from 8.4 seconds to 8.8 seconds on 1G lguest->Host TCP xmit. So now we only do that as a fallback. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-16 06:15:10 +08:00
if (!skb)
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
return ERR_PTR(err);
tun: fallback if skb_alloc() fails on big packets skb_alloc produces linear packets (using kmalloc()). That can fail, so should we fall back to making paged skbs. My original version of this patch always allocate paged skbs for big packets. But that made performance drop from 8.4 seconds to 8.8 seconds on 1G lguest->Host TCP xmit. So now we only do that as a fallback. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-16 06:15:10 +08:00
skb_reserve(skb, prepad);
skb_put(skb, linear);
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
skb->data_len = len - linear;
skb->len += len - linear;
tun: fallback if skb_alloc() fails on big packets skb_alloc produces linear packets (using kmalloc()). That can fail, so should we fall back to making paged skbs. My original version of this patch always allocate paged skbs for big packets. But that made performance drop from 8.4 seconds to 8.8 seconds on 1G lguest->Host TCP xmit. So now we only do that as a fallback. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-16 06:15:10 +08:00
return skb;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* Get packet from user space buffer */
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
static __inline__ ssize_t tun_get_user(struct tun_struct *tun,
tun: fix tun_chr_aio_write so that aio works aio_write gets const struct iovec * but tun_chr_aio_write casts this to struct iovec * and modifies the iovec. As a result, attempts to use io_submit to send packets to a tun device fail with weird errors such as EINVAL. Since tun is the only user of skb_copy_datagram_from_iovec, we can fix this simply by changing the later so that it does not touch the iovec passed to it. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 09:26:11 +08:00
const struct iovec *iv, size_t count,
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
int noblock)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
net: replace uses of __constant_{endian} Base versions handle constant folding now. Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-01 16:45:17 +08:00
struct tun_pi pi = { 0, cpu_to_be16(ETH_P_IP) };
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
struct sk_buff *skb;
size_t len = count, align = 0;
tun: Allow GSO using virtio_net_hdr Add a IFF_VNET_HDR flag. This uses the same ABI as virtio_net (ie. prepending struct virtio_net_hdr to packets) to indicate GSO and checksum information. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-03 18:48:02 +08:00
struct virtio_net_hdr gso = { 0 };
tun: fix tun_chr_aio_write so that aio works aio_write gets const struct iovec * but tun_chr_aio_write casts this to struct iovec * and modifies the iovec. As a result, attempts to use io_submit to send packets to a tun device fail with weird errors such as EINVAL. Since tun is the only user of skb_copy_datagram_from_iovec, we can fix this simply by changing the later so that it does not touch the iovec passed to it. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 09:26:11 +08:00
int offset = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (!(tun->flags & TUN_NO_PI)) {
if ((len -= sizeof(pi)) > count)
return -EINVAL;
tun: fix tun_chr_aio_write so that aio works aio_write gets const struct iovec * but tun_chr_aio_write casts this to struct iovec * and modifies the iovec. As a result, attempts to use io_submit to send packets to a tun device fail with weird errors such as EINVAL. Since tun is the only user of skb_copy_datagram_from_iovec, we can fix this simply by changing the later so that it does not touch the iovec passed to it. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 09:26:11 +08:00
if (memcpy_fromiovecend((void *)&pi, iv, 0, sizeof(pi)))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return -EFAULT;
tun: fix tun_chr_aio_write so that aio works aio_write gets const struct iovec * but tun_chr_aio_write casts this to struct iovec * and modifies the iovec. As a result, attempts to use io_submit to send packets to a tun device fail with weird errors such as EINVAL. Since tun is the only user of skb_copy_datagram_from_iovec, we can fix this simply by changing the later so that it does not touch the iovec passed to it. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 09:26:11 +08:00
offset += sizeof(pi);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
tun: Allow GSO using virtio_net_hdr Add a IFF_VNET_HDR flag. This uses the same ABI as virtio_net (ie. prepending struct virtio_net_hdr to packets) to indicate GSO and checksum information. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-03 18:48:02 +08:00
if (tun->flags & TUN_VNET_HDR) {
if ((len -= sizeof(gso)) > count)
return -EINVAL;
tun: fix tun_chr_aio_write so that aio works aio_write gets const struct iovec * but tun_chr_aio_write casts this to struct iovec * and modifies the iovec. As a result, attempts to use io_submit to send packets to a tun device fail with weird errors such as EINVAL. Since tun is the only user of skb_copy_datagram_from_iovec, we can fix this simply by changing the later so that it does not touch the iovec passed to it. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 09:26:11 +08:00
if (memcpy_fromiovecend((void *)&gso, iv, offset, sizeof(gso)))
tun: Allow GSO using virtio_net_hdr Add a IFF_VNET_HDR flag. This uses the same ABI as virtio_net (ie. prepending struct virtio_net_hdr to packets) to indicate GSO and checksum information. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-03 18:48:02 +08:00
return -EFAULT;
if (gso.hdr_len > len)
return -EINVAL;
tun: fix tun_chr_aio_write so that aio works aio_write gets const struct iovec * but tun_chr_aio_write casts this to struct iovec * and modifies the iovec. As a result, attempts to use io_submit to send packets to a tun device fail with weird errors such as EINVAL. Since tun is the only user of skb_copy_datagram_from_iovec, we can fix this simply by changing the later so that it does not touch the iovec passed to it. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 09:26:11 +08:00
offset += sizeof(pi);
tun: Allow GSO using virtio_net_hdr Add a IFF_VNET_HDR flag. This uses the same ABI as virtio_net (ie. prepending struct virtio_net_hdr to packets) to indicate GSO and checksum information. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-03 18:48:02 +08:00
}
net: check for underlength tap writes If the user gives a packet under 14 bytes, we'll end up reading off the end of the skb (not oopsing, just reading off the end). Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyanskiy <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-13 09:49:30 +08:00
if ((tun->flags & TUN_TYPE_MASK) == TUN_TAP_DEV) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
align = NET_IP_ALIGN;
tun: Fix crash with non-GSO users When I made the tun driver use non-linear packets as the preferred option, it broke non-GSO users because they would end up allocating a completely non-linear packet, which triggers a crash when we call eth_type_trans. This patch reverts non-GSO users to using linear packets and adds a check to ensure that GSO users can't cause crashes in the same way. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-14 17:09:43 +08:00
if (unlikely(len < ETH_HLEN ||
(gso.hdr_len && gso.hdr_len < ETH_HLEN)))
net: check for underlength tap writes If the user gives a packet under 14 bytes, we'll end up reading off the end of the skb (not oopsing, just reading off the end). Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyanskiy <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-13 09:49:30 +08:00
return -EINVAL;
}
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
skb = tun_alloc_skb(tun, align, len, gso.hdr_len, noblock);
if (IS_ERR(skb)) {
if (PTR_ERR(skb) != -EAGAIN)
tun->dev->stats.rx_dropped++;
return PTR_ERR(skb);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
tun: fix tun_chr_aio_write so that aio works aio_write gets const struct iovec * but tun_chr_aio_write casts this to struct iovec * and modifies the iovec. As a result, attempts to use io_submit to send packets to a tun device fail with weird errors such as EINVAL. Since tun is the only user of skb_copy_datagram_from_iovec, we can fix this simply by changing the later so that it does not touch the iovec passed to it. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 09:26:11 +08:00
if (skb_copy_datagram_from_iovec(skb, 0, iv, offset, len)) {
[NET] drivers/net: statistics cleanup #1 -- save memory and shrink code We now have struct net_device_stats embedded in struct net_device, and the default ->get_stats() hook does the obvious thing for us. Run through drivers/net/* and remove the driver-local storage of statistics, and driver-local ->get_stats() hook where applicable. This was just the low-hanging fruit in drivers/net; plenty more drivers remain to be updated. [ Resolved conflicts with napi_struct changes and fix sunqe build regression... -DaveM ] Signed-off-by: Jeff Garzik <jeff@garzik.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 08:41:50 +08:00
tun->dev->stats.rx_dropped++;
[TUN]: Fix leak in tun_get_user() We're leaking an skb in a failure path in this function. Coverity #632 Signed-off-by: Dave Jones <davej@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-03-12 10:49:13 +08:00
kfree_skb(skb);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return -EFAULT;
[TUN]: Fix leak in tun_get_user() We're leaking an skb in a failure path in this function. Coverity #632 Signed-off-by: Dave Jones <davej@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-03-12 10:49:13 +08:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
tun: Allow GSO using virtio_net_hdr Add a IFF_VNET_HDR flag. This uses the same ABI as virtio_net (ie. prepending struct virtio_net_hdr to packets) to indicate GSO and checksum information. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-03 18:48:02 +08:00
if (gso.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) {
if (!skb_partial_csum_set(skb, gso.csum_start,
gso.csum_offset)) {
tun->dev->stats.rx_frame_errors++;
kfree_skb(skb);
return -EINVAL;
}
} else if (tun->flags & TUN_NOCHECKSUM)
skb->ip_summed = CHECKSUM_UNNECESSARY;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
switch (tun->flags & TUN_TYPE_MASK) {
case TUN_TUN_DEV:
tun: Proper handling of IPv6 header in tun driver when TUN_NO_PI is set By default, tun.c running in TUN_TUN_DEV mode will set the protocol of packet to IPv4 if TUN_NO_PI is set. My program failed to work when I assumed that the driver will check the first nibble of packet, determine IP version and set the appropriate protocol. Signed-off-by: Ang Way Chuang <wcang@nav6.org> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-06-18 12:10:33 +08:00
if (tun->flags & TUN_NO_PI) {
switch (skb->data[0] & 0xf0) {
case 0x40:
pi.proto = htons(ETH_P_IP);
break;
case 0x60:
pi.proto = htons(ETH_P_IPV6);
break;
default:
tun->dev->stats.rx_dropped++;
kfree_skb(skb);
return -EINVAL;
}
}
[SK_BUFF]: Introduce skb_reset_mac_header(skb) For the common, open coded 'skb->mac.raw = skb->data' operation, so that we can later turn skb->mac.raw into a offset, reducing the size of struct sk_buff in 64bit land while possibly keeping it as a pointer on 32bit. This one touches just the most simple case, next will handle the slightly more "complex" cases. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-03-20 06:30:44 +08:00
skb_reset_mac_header(skb);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
skb->protocol = pi.proto;
[ETH]: Make eth_type_trans set skb->dev like the other *_type_trans One less thing for drivers writers to worry about. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-04-26 08:40:23 +08:00
skb->dev = tun->dev;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
break;
case TUN_TAP_DEV:
skb->protocol = eth_type_trans(skb, tun->dev);
break;
};
tun: Allow GSO using virtio_net_hdr Add a IFF_VNET_HDR flag. This uses the same ABI as virtio_net (ie. prepending struct virtio_net_hdr to packets) to indicate GSO and checksum information. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-03 18:48:02 +08:00
if (gso.gso_type != VIRTIO_NET_HDR_GSO_NONE) {
pr_debug("GSO!\n");
switch (gso.gso_type & ~VIRTIO_NET_HDR_GSO_ECN) {
case VIRTIO_NET_HDR_GSO_TCPV4:
skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4;
break;
case VIRTIO_NET_HDR_GSO_TCPV6:
skb_shinfo(skb)->gso_type = SKB_GSO_TCPV6;
break;
default:
tun->dev->stats.rx_frame_errors++;
kfree_skb(skb);
return -EINVAL;
}
if (gso.gso_type & VIRTIO_NET_HDR_GSO_ECN)
skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_ECN;
skb_shinfo(skb)->gso_size = gso.gso_size;
if (skb_shinfo(skb)->gso_size == 0) {
tun->dev->stats.rx_frame_errors++;
kfree_skb(skb);
return -EINVAL;
}
/* Header must be checked, and gso_segs computed. */
skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY;
skb_shinfo(skb)->gso_segs = 0;
}
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
netif_rx_ni(skb);
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
[NET] drivers/net: statistics cleanup #1 -- save memory and shrink code We now have struct net_device_stats embedded in struct net_device, and the default ->get_stats() hook does the obvious thing for us. Run through drivers/net/* and remove the driver-local storage of statistics, and driver-local ->get_stats() hook where applicable. This was just the low-hanging fruit in drivers/net; plenty more drivers remain to be updated. [ Resolved conflicts with napi_struct changes and fix sunqe build regression... -DaveM ] Signed-off-by: Jeff Garzik <jeff@garzik.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 08:41:50 +08:00
tun->dev->stats.rx_packets++;
tun->dev->stats.rx_bytes += len;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return count;
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
[PATCH] Remove readv/writev methods and use aio_read/aio_write instead This patch removes readv() and writev() methods and replaces them with aio_read()/aio_write() methods. Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-01 14:28:47 +08:00
static ssize_t tun_chr_aio_write(struct kiocb *iocb, const struct iovec *iv,
unsigned long count, loff_t pos)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
struct file *file = iocb->ki_filp;
tun: Fix merge error When forward-porting the tun accounting patch I managed to break the send path compltely by dropping the tun_get call. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-15 12:46:39 +08:00
struct tun_struct *tun = tun_get(file);
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
ssize_t result;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (!tun)
return -EBADFD;
DBG(KERN_INFO "%s: tun_chr_write %ld\n", tun->dev->name, count);
tun: fix tun_chr_aio_write so that aio works aio_write gets const struct iovec * but tun_chr_aio_write casts this to struct iovec * and modifies the iovec. As a result, attempts to use io_submit to send packets to a tun device fail with weird errors such as EINVAL. Since tun is the only user of skb_copy_datagram_from_iovec, we can fix this simply by changing the later so that it does not touch the iovec passed to it. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 09:26:11 +08:00
result = tun_get_user(tun, iv, iov_length(iv, count),
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
file->f_flags & O_NONBLOCK);
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
tun_put(tun);
return result;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
/* Put packet to the user space buffer */
static __inline__ ssize_t tun_put_user(struct tun_struct *tun,
struct sk_buff *skb,
tun: fix tun_chr_aio_read so that aio works aio_read gets const struct iovec * but tun_chr_aio_read casts this to struct iovec * and modifies the iovec. As a result, attempts to use io_submit to get packets from a tun device fail with weird errors such as EINVAL. Fix by using the new skb_copy_datagram_const_iovec. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 09:25:59 +08:00
const struct iovec *iv, int len)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
struct tun_pi pi = { 0, skb->protocol };
ssize_t total = 0;
if (!(tun->flags & TUN_NO_PI)) {
if ((len -= sizeof(pi)) < 0)
return -EINVAL;
if (len < skb->len) {
/* Packet will be striped */
pi.flags |= TUN_PKT_STRIP;
}
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
tun: fix tun_chr_aio_read so that aio works aio_read gets const struct iovec * but tun_chr_aio_read casts this to struct iovec * and modifies the iovec. As a result, attempts to use io_submit to get packets from a tun device fail with weird errors such as EINVAL. Fix by using the new skb_copy_datagram_const_iovec. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 09:25:59 +08:00
if (memcpy_toiovecend(iv, (void *) &pi, 0, sizeof(pi)))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return -EFAULT;
total += sizeof(pi);
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
tun: Allow GSO using virtio_net_hdr Add a IFF_VNET_HDR flag. This uses the same ABI as virtio_net (ie. prepending struct virtio_net_hdr to packets) to indicate GSO and checksum information. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-03 18:48:02 +08:00
if (tun->flags & TUN_VNET_HDR) {
struct virtio_net_hdr gso = { 0 }; /* no info leak */
if ((len -= sizeof(gso)) < 0)
return -EINVAL;
if (skb_is_gso(skb)) {
struct skb_shared_info *sinfo = skb_shinfo(skb);
/* This is a hint as to how much should be linear. */
gso.hdr_len = skb_headlen(skb);
gso.gso_size = sinfo->gso_size;
if (sinfo->gso_type & SKB_GSO_TCPV4)
gso.gso_type = VIRTIO_NET_HDR_GSO_TCPV4;
else if (sinfo->gso_type & SKB_GSO_TCPV6)
gso.gso_type = VIRTIO_NET_HDR_GSO_TCPV6;
else
BUG();
if (sinfo->gso_type & SKB_GSO_TCP_ECN)
gso.gso_type |= VIRTIO_NET_HDR_GSO_ECN;
} else
gso.gso_type = VIRTIO_NET_HDR_GSO_NONE;
if (skb->ip_summed == CHECKSUM_PARTIAL) {
gso.flags = VIRTIO_NET_HDR_F_NEEDS_CSUM;
gso.csum_start = skb->csum_start - skb_headroom(skb);
gso.csum_offset = skb->csum_offset;
} /* else everything is zero */
tun: fix tun_chr_aio_read so that aio works aio_read gets const struct iovec * but tun_chr_aio_read casts this to struct iovec * and modifies the iovec. As a result, attempts to use io_submit to get packets from a tun device fail with weird errors such as EINVAL. Fix by using the new skb_copy_datagram_const_iovec. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 09:25:59 +08:00
if (unlikely(memcpy_toiovecend(iv, (void *)&gso, total,
sizeof(gso))))
tun: Allow GSO using virtio_net_hdr Add a IFF_VNET_HDR flag. This uses the same ABI as virtio_net (ie. prepending struct virtio_net_hdr to packets) to indicate GSO and checksum information. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-03 18:48:02 +08:00
return -EFAULT;
total += sizeof(gso);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
len = min_t(int, skb->len, len);
tun: fix tun_chr_aio_read so that aio works aio_read gets const struct iovec * but tun_chr_aio_read casts this to struct iovec * and modifies the iovec. As a result, attempts to use io_submit to get packets from a tun device fail with weird errors such as EINVAL. Fix by using the new skb_copy_datagram_const_iovec. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 09:25:59 +08:00
skb_copy_datagram_const_iovec(skb, 0, iv, total, len);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
total += len;
[NET] drivers/net: statistics cleanup #1 -- save memory and shrink code We now have struct net_device_stats embedded in struct net_device, and the default ->get_stats() hook does the obvious thing for us. Run through drivers/net/* and remove the driver-local storage of statistics, and driver-local ->get_stats() hook where applicable. This was just the low-hanging fruit in drivers/net; plenty more drivers remain to be updated. [ Resolved conflicts with napi_struct changes and fix sunqe build regression... -DaveM ] Signed-off-by: Jeff Garzik <jeff@garzik.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 08:41:50 +08:00
tun->dev->stats.tx_packets++;
tun->dev->stats.tx_bytes += len;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return total;
}
[PATCH] Remove readv/writev methods and use aio_read/aio_write instead This patch removes readv() and writev() methods and replaces them with aio_read()/aio_write() methods. Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-01 14:28:47 +08:00
static ssize_t tun_chr_aio_read(struct kiocb *iocb, const struct iovec *iv,
unsigned long count, loff_t pos)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
[PATCH] Remove readv/writev methods and use aio_read/aio_write instead This patch removes readv() and writev() methods and replaces them with aio_read()/aio_write() methods. Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-01 14:28:47 +08:00
struct file *file = iocb->ki_filp;
tun: Move read_wait into tun_file The poll interface requires that the waitqueue exist while the struct file is open. In the rare case when a tun device disappears before the tun file closes we fail to provide this property, so move read_wait. This is safe now that tun_net_xmit is atomic with tun_detach. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:03:21 +08:00
struct tun_file *tfile = file->private_data;
struct tun_struct *tun = __tun_get(tfile);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
DECLARE_WAITQUEUE(wait, current);
struct sk_buff *skb;
ssize_t len, ret = 0;
if (!tun)
return -EBADFD;
DBG(KERN_INFO "%s: tun_chr_read\n", tun->dev->name);
[TUN]: Use iov_length() Use iov_length() instead of tun's homemade iov_total(). Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-11-20 14:46:51 +08:00
len = iov_length(iv, count);
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
if (len < 0) {
ret = -EINVAL;
goto out;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
tun: Fix sk_sleep races when attaching/detaching As the sk_sleep wait queue actually lives in tfile, which may be detached from the tun device, bad things will happen when we use sk_sleep after detaching. Since the tun device is the persistent data structure here (when requested by the user), it makes much more sense to have the wait queue live there. There is no reason to have it in tfile at all since the only time we can wait is if we have a tun attached. In fact we already have a wait queue in tun_struct, so we might as well use it. Reported-by: Eric W. Biederman <ebiederm@xmission.com> Tested-by: Christian Borntraeger <borntraeger@de.ibm.com> Tested-by: Patrick McHardy <kaber@trash.net> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 06:35:50 +08:00
add_wait_queue(&tun->socket.wait, &wait);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
while (len) {
current->state = TASK_INTERRUPTIBLE;
/* Read frames from the queue */
if (!(skb=skb_dequeue(&tun->readq))) {
if (file->f_flags & O_NONBLOCK) {
ret = -EAGAIN;
break;
}
if (signal_pending(current)) {
ret = -ERESTARTSYS;
break;
}
tun: Fix races between tun_net_close and free_netdev. The tun code does not cope gracefully if the network device goes away before the tun file descriptor is closed. It looks like we can trigger this with rmmod, and moving tun devices between network namespaces will allow this to be triggered when network namespaces exit. To fix this I introduce an intermediate data structure tun_file which holds a count of users and a pointer to the struct tun_struct. tun_get increments that reference count if it is greater than 0. tun_put decrements that reference count and detaches from the network device if the count is 0. While we have a file attached to the network device I hold a reference to the network device keeping it from going away completely. When a network device is unregistered I decrement the count of the attached tun_file and if that was the last user I detach the tun_file, and all processes on read_wait are woken up to ensure they do not sleep indefinitely. As some of those sleeps happen with the count on the tun device elevated waking up the read waiters ensures that tun_file will be detached in a timely manner. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:07:17 +08:00
if (tun->dev->reg_state != NETREG_REGISTERED) {
ret = -EIO;
break;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* Nothing to read, let's sleep */
schedule();
continue;
}
netif_wake_queue(tun->dev);
tun: fix tun_chr_aio_read so that aio works aio_read gets const struct iovec * but tun_chr_aio_read casts this to struct iovec * and modifies the iovec. As a result, attempts to use io_submit to get packets from a tun device fail with weird errors such as EINVAL. Fix by using the new skb_copy_datagram_const_iovec. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 09:25:59 +08:00
ret = tun_put_user(tun, skb, iv, len);
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
kfree_skb(skb);
break;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
current->state = TASK_RUNNING;
tun: Fix sk_sleep races when attaching/detaching As the sk_sleep wait queue actually lives in tfile, which may be detached from the tun device, bad things will happen when we use sk_sleep after detaching. Since the tun device is the persistent data structure here (when requested by the user), it makes much more sense to have the wait queue live there. There is no reason to have it in tfile at all since the only time we can wait is if we have a tun attached. In fact we already have a wait queue in tun_struct, so we might as well use it. Reported-by: Eric W. Biederman <ebiederm@xmission.com> Tested-by: Christian Borntraeger <borntraeger@de.ibm.com> Tested-by: Patrick McHardy <kaber@trash.net> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 06:35:50 +08:00
remove_wait_queue(&tun->socket.wait, &wait);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
out:
tun_put(tun);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return ret;
}
static void tun_setup(struct net_device *dev)
{
struct tun_struct *tun = netdev_priv(dev);
skb_queue_head_init(&tun->readq);
tun->owner = -1;
[NET]: Allow group ownership of TUN/TAP devices. Introduce a new syscall TUNSETGROUP for group ownership setting of tap devices. The user now is allowed to send packages if either his euid or his egid matches the one specified via tunctl (via -u or -g respecitvely). If both, gid and uid, are set via tunctl, both have to match. Signed-off-by: Guido Guenther <agx@sigxcpu.org> Signed-off-by: Jeff Dike <jdike@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-07-03 13:50:25 +08:00
tun->group = -1;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
dev->ethtool_ops = &tun_ethtool_ops;
tun: Only free a netdev when all tun descriptors are closed The commit c70f182940f988448f3c12a209d18b1edc276e33 ("tun: Fix races between tun_net_close and free_netdev") fixed a race where an asynchronous deletion of a tun device can hose a poll(2) on a tun fd attached to that device. However, this came at the cost of moving the tun wait queue into the tun file data structure. The problem with this is that it imposes restrictions on when and where the tun device can access the wait queue since the tun file may change at any time due to detaching and reattaching. In particular, now that we need to use the wait queue on the receive path it becomes difficult to properly synchronise this with the detachment of the tun device. This patch solves the original race in a different way. Since the race is only because the underlying memory gets freed, we can prevent it simply by ensuring that we don't do that until all tun descriptors ever attached to the device (even if they have since be detached because they may still be sitting in poll) have been closed. This is done by using reference counting the attached tun file descriptors. The refcount in tun->sk has been reappropriated for this purpose since it was already being used for that, albeit from the opposite angle. Note that we no longer zero tfile->tun since tun_get will return NULL anyway after the refcount on tfile hits zero. Instead it represents whether this device has ever been attached to a device. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-18 22:15:52 +08:00
dev->destructor = tun_free_netdev;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
tun: Implement ip link del tunXXX This greatly simplifies testing to verify I have fixed the problems with a tun device disappearing when the tun file descriptor is still held open. Further it allows removal network namespace operations for the tun driver. Reducing the network namespace handling in the driver to the minimum. i.e. When we are creating a tun device. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-22 08:02:16 +08:00
/* Trivial set of netlink ops to allow deleting tun or tap
* device with netlink.
*/
static int tun_validate(struct nlattr *tb[], struct nlattr *data[])
{
return -EINVAL;
}
static struct rtnl_link_ops tun_link_ops __read_mostly = {
.kind = DRV_NAME,
.priv_size = sizeof(struct tun_struct),
.setup = tun_setup,
.validate = tun_validate,
};
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
static void tun_sock_write_space(struct sock *sk)
{
struct tun_struct *tun;
if (!sock_writeable(sk))
return;
if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
wake_up_interruptible_sync(sk->sk_sleep);
if (!test_and_clear_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags))
return;
tun = container_of(sk, struct tun_sock, sk)->tun;
kill_fasync(&tun->fasync, SIGIO, POLL_OUT);
}
static void tun_sock_destruct(struct sock *sk)
{
tun: Only free a netdev when all tun descriptors are closed The commit c70f182940f988448f3c12a209d18b1edc276e33 ("tun: Fix races between tun_net_close and free_netdev") fixed a race where an asynchronous deletion of a tun device can hose a poll(2) on a tun fd attached to that device. However, this came at the cost of moving the tun wait queue into the tun file data structure. The problem with this is that it imposes restrictions on when and where the tun device can access the wait queue since the tun file may change at any time due to detaching and reattaching. In particular, now that we need to use the wait queue on the receive path it becomes difficult to properly synchronise this with the detachment of the tun device. This patch solves the original race in a different way. Since the race is only because the underlying memory gets freed, we can prevent it simply by ensuring that we don't do that until all tun descriptors ever attached to the device (even if they have since be detached because they may still be sitting in poll) have been closed. This is done by using reference counting the attached tun file descriptors. The refcount in tun->sk has been reappropriated for this purpose since it was already being used for that, albeit from the opposite angle. Note that we no longer zero tfile->tun since tun_get will return NULL anyway after the refcount on tfile hits zero. Instead it represents whether this device has ever been attached to a device. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-18 22:15:52 +08:00
free_netdev(container_of(sk, struct tun_sock, sk)->tun->dev);
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
}
static struct proto tun_proto = {
.name = "tun",
.owner = THIS_MODULE,
.obj_size = sizeof(struct tun_sock),
};
tun: Implement ip link del tunXXX This greatly simplifies testing to verify I have fixed the problems with a tun device disappearing when the tun file descriptor is still held open. Further it allows removal network namespace operations for the tun driver. Reducing the network namespace handling in the driver to the minimum. i.e. When we are creating a tun device. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-22 08:02:16 +08:00
[TUN]: Make the tun_dev_list per-net. Remove the static tun_dev_list and replace its occurrences in driver with per-net one. It is used in two places - in tun_set_iff and tun_cleanup. In the first case it's legal to use current net_ns. In the cleanup call - move the loop, that unregisters all devices in net exit hook. Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-16 15:41:16 +08:00
static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
struct sock *sk;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
struct tun_struct *tun;
struct net_device *dev;
int err;
tun: Remove unnecessary tun_get_by_name Currently the tun driver keeps a private list of tun devices for what appears to be a small gain in performance when reconnecting a file descriptor to an existing tun or tap device. So simplify the code by removing it. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 18:56:20 +08:00
dev = __dev_get_by_name(net, ifr->ifr_name);
if (dev) {
if ((ifr->ifr_flags & IFF_TUN) && dev->netdev_ops == &tun_netdev_ops)
tun = netdev_priv(dev);
else if ((ifr->ifr_flags & IFF_TAP) && dev->netdev_ops == &tap_netdev_ops)
tun = netdev_priv(dev);
else
return -EINVAL;
tun: Fix races in tun_set_iff It is possible for two different tasks with access to the same file descriptor to call tun_set_iff on it at the same time and race to attach to a tap device. Prevent this by placing all of the logic to attach to a file descriptor in one function and testing the file descriptor to be certain it is not already attached to another tun device. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 18:57:48 +08:00
err = tun_attach(tun, file);
if (err < 0)
return err;
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
else {
char *name;
unsigned long flags = 0;
err = -EINVAL;
[NET]: Require CAP_NET_ADMIN to create tuntap devices. The tuntap driver allows an admin to create persistent devices and assign ownership of them to individual users. Unfortunately, relaxing the permissions on the /dev/net/tun device node so that they can actually use those devices will _also_ allow those users to create arbitrary new devices of their own. This patch corrects that, and adjusts the recommended permissions for the device node accordingly. Signed-off-By: David Woodhouse <dwmw2@infradead.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-23 07:07:52 +08:00
if (!capable(CAP_NET_ADMIN))
return -EPERM;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* Set dev type */
if (ifr->ifr_flags & IFF_TUN) {
/* TUN device */
flags |= TUN_TUN_DEV;
name = "tun%d";
} else if (ifr->ifr_flags & IFF_TAP) {
/* TAP device */
flags |= TUN_TAP_DEV;
name = "tap%d";
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
} else
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
goto failed;
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (*ifr->ifr_name)
name = ifr->ifr_name;
dev = alloc_netdev(sizeof(struct tun_struct), name,
tun_setup);
if (!dev)
return -ENOMEM;
[TUN]: Allow to register tun devices in namespace. This is basically means that a net is set for a new device, but actually also involves two more steps: 1. mark the tun device as "local", i.e. do not allow for it to move across namespaces. This is done so, since tun device is most often associated to some file (and thus to some process) and moving the device alone is not valid while keeping the file and the process outside. The need in ability to move a detached persistent device is to be investigated later. 2. get the tun device's net when tun becomes attached and put one when it becomes detached. This is needed to handle the case when a task owning the tun dies, but a files lives for some more time - in this case we must not allow for net to be freed, since its exit hook will spoil that file's private data by unregistering the tun from under tun_chr_close. Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-16 15:41:53 +08:00
dev_net_set(dev, net);
tun: Implement ip link del tunXXX This greatly simplifies testing to verify I have fixed the problems with a tun device disappearing when the tun file descriptor is still held open. Further it allows removal network namespace operations for the tun driver. Reducing the network namespace handling in the driver to the minimum. i.e. When we are creating a tun device. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-22 08:02:16 +08:00
dev->rtnl_link_ops = &tun_link_ops;
tun: convert to net_device_ops Convert the TUN/TAP tunnel driver to net_device_ops. Split the ops in two, and retain compatability. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 14:10:37 +08:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
tun = netdev_priv(dev);
tun->dev = dev;
tun->flags = flags;
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
tun->txflt.count = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
err = -ENOMEM;
sk = sk_alloc(net, AF_UNSPEC, GFP_KERNEL, &tun_proto);
if (!sk)
goto err_free_dev;
tun: Fix sk_sleep races when attaching/detaching As the sk_sleep wait queue actually lives in tfile, which may be detached from the tun device, bad things will happen when we use sk_sleep after detaching. Since the tun device is the persistent data structure here (when requested by the user), it makes much more sense to have the wait queue live there. There is no reason to have it in tfile at all since the only time we can wait is if we have a tun attached. In fact we already have a wait queue in tun_struct, so we might as well use it. Reported-by: Eric W. Biederman <ebiederm@xmission.com> Tested-by: Christian Borntraeger <borntraeger@de.ibm.com> Tested-by: Patrick McHardy <kaber@trash.net> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-20 06:35:50 +08:00
init_waitqueue_head(&tun->socket.wait);
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
sock_init_data(&tun->socket, sk);
sk->sk_write_space = tun_sock_write_space;
sk->sk_sndbuf = INT_MAX;
tun->sk = sk;
container_of(sk, struct tun_sock, sk)->tun = tun;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
tun_net_init(dev);
if (strchr(dev->name, '%')) {
err = dev_alloc_name(dev, dev->name);
if (err < 0)
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
goto err_free_sk;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
err = -EINVAL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
err = register_netdevice(tun->dev);
if (err < 0)
tun: Only free a netdev when all tun descriptors are closed The commit c70f182940f988448f3c12a209d18b1edc276e33 ("tun: Fix races between tun_net_close and free_netdev") fixed a race where an asynchronous deletion of a tun device can hose a poll(2) on a tun fd attached to that device. However, this came at the cost of moving the tun wait queue into the tun file data structure. The problem with this is that it imposes restrictions on when and where the tun device can access the wait queue since the tun file may change at any time due to detaching and reattaching. In particular, now that we need to use the wait queue on the receive path it becomes difficult to properly synchronise this with the detachment of the tun device. This patch solves the original race in a different way. Since the race is only because the underlying memory gets freed, we can prevent it simply by ensuring that we don't do that until all tun descriptors ever attached to the device (even if they have since be detached because they may still be sitting in poll) have been closed. This is done by using reference counting the attached tun file descriptors. The refcount in tun->sk has been reappropriated for this purpose since it was already being used for that, albeit from the opposite angle. Note that we no longer zero tfile->tun since tun_get will return NULL anyway after the refcount on tfile hits zero. Instead it represents whether this device has ever been attached to a device. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-18 22:15:52 +08:00
goto err_free_sk;
sk->sk_destruct = tun_sock_destruct;
tun: Fix races in tun_set_iff It is possible for two different tasks with access to the same file descriptor to call tun_set_iff on it at the same time and race to attach to a tap device. Prevent this by placing all of the logic to attach to a file descriptor in one function and testing the file descriptor to be certain it is not already attached to another tun device. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 18:57:48 +08:00
err = tun_attach(tun, file);
if (err < 0)
tun: Only free a netdev when all tun descriptors are closed The commit c70f182940f988448f3c12a209d18b1edc276e33 ("tun: Fix races between tun_net_close and free_netdev") fixed a race where an asynchronous deletion of a tun device can hose a poll(2) on a tun fd attached to that device. However, this came at the cost of moving the tun wait queue into the tun file data structure. The problem with this is that it imposes restrictions on when and where the tun device can access the wait queue since the tun file may change at any time due to detaching and reattaching. In particular, now that we need to use the wait queue on the receive path it becomes difficult to properly synchronise this with the detachment of the tun device. This patch solves the original race in a different way. Since the race is only because the underlying memory gets freed, we can prevent it simply by ensuring that we don't do that until all tun descriptors ever attached to the device (even if they have since be detached because they may still be sitting in poll) have been closed. This is done by using reference counting the attached tun file descriptors. The refcount in tun->sk has been reappropriated for this purpose since it was already being used for that, albeit from the opposite angle. Note that we no longer zero tfile->tun since tun_get will return NULL anyway after the refcount on tfile hits zero. Instead it represents whether this device has ever been attached to a device. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-18 22:15:52 +08:00
goto failed;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
DBG(KERN_INFO "%s: tun_set_iff\n", tun->dev->name);
if (ifr->ifr_flags & IFF_NO_PI)
tun->flags |= TUN_NO_PI;
tun: impossible to deassert IFF_ONE_QUEUE or IFF_NO_PI From: "Nathaniel Filardo" <nwfilardo@gmail.com> Taken from http://bugzilla.kernel.org/show_bug.cgi?id=9806 The TUN/TAP driver only permits one-way transitions of IFF_NO_PI or IFF_ONE_QUEUE during the lifetime of a tap/tun interface. Note that tun_set_iff contains 541 if (ifr->ifr_flags & IFF_NO_PI) 542 tun->flags |= TUN_NO_PI; 543 544 if (ifr->ifr_flags & IFF_ONE_QUEUE) 545 tun->flags |= TUN_ONE_QUEUE; This is easily fixed by adding else branches which clear these bits. Steps to reproduce: This is easily reproduced by setting an interface persistant using tunctl then attempting to open it as IFF_TAP or IFF_TUN, without asserting the IFF_NO_PI flag. The ioctl() will succeed and the ifr.flags word is not modified, but the interface remains in IFF_NO_PI mode (as it was set by tunctl). Acked-by: Maxim Krasnyansky <maxk@qualcomm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-02-05 19:05:07 +08:00
else
tun->flags &= ~TUN_NO_PI;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (ifr->ifr_flags & IFF_ONE_QUEUE)
tun->flags |= TUN_ONE_QUEUE;
tun: impossible to deassert IFF_ONE_QUEUE or IFF_NO_PI From: "Nathaniel Filardo" <nwfilardo@gmail.com> Taken from http://bugzilla.kernel.org/show_bug.cgi?id=9806 The TUN/TAP driver only permits one-way transitions of IFF_NO_PI or IFF_ONE_QUEUE during the lifetime of a tap/tun interface. Note that tun_set_iff contains 541 if (ifr->ifr_flags & IFF_NO_PI) 542 tun->flags |= TUN_NO_PI; 543 544 if (ifr->ifr_flags & IFF_ONE_QUEUE) 545 tun->flags |= TUN_ONE_QUEUE; This is easily fixed by adding else branches which clear these bits. Steps to reproduce: This is easily reproduced by setting an interface persistant using tunctl then attempting to open it as IFF_TAP or IFF_TUN, without asserting the IFF_NO_PI flag. The ioctl() will succeed and the ifr.flags word is not modified, but the interface remains in IFF_NO_PI mode (as it was set by tunctl). Acked-by: Maxim Krasnyansky <maxk@qualcomm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-02-05 19:05:07 +08:00
else
tun->flags &= ~TUN_ONE_QUEUE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
tun: Allow GSO using virtio_net_hdr Add a IFF_VNET_HDR flag. This uses the same ABI as virtio_net (ie. prepending struct virtio_net_hdr to packets) to indicate GSO and checksum information. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-03 18:48:02 +08:00
if (ifr->ifr_flags & IFF_VNET_HDR)
tun->flags |= TUN_VNET_HDR;
else
tun->flags &= ~TUN_VNET_HDR;
tun: Persistent devices can get stuck in xoff state The scenario goes like this. App stops reading from tun/tap. TX queue gets full and driver does netif_stop_queue(). App closes fd and TX queue gets flushed as part of the cleanup. Next time the app opens tun/tap and starts reading from it but the xoff state is not cleared. We're stuck. Normally xoff state is cleared when netdev is brought up. But in the case of persistent devices this happens only during initial setup. The fix is trivial. If device is already up when an app opens it we clear xoff state and that gets things moving again. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Tested-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-11 07:59:11 +08:00
/* Make sure persistent devices do not get stuck in
* xoff state.
*/
if (netif_running(tun->dev))
netif_wake_queue(tun->dev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
strcpy(ifr->ifr_name, tun->dev->name);
return 0;
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
err_free_sk:
sock_put(sk);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
err_free_dev:
free_netdev(dev);
failed:
return err;
}
tun: TUNGETIFF interface to query name and flags Add a TUNGETIFF interface so that userspace can query a tun/tap descriptor for its name and flags. This is needed because it is common for one app to create a tap interface, exec another app and pass it the file descriptor for the interface. Without TUNGETIFF the spawned app has no way of detecting wheter the interface has e.g. IFF_VNET_HDR set. Signed-off-by: Mark McLoughlin <markmc@redhat.com> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-16 06:09:56 +08:00
static int tun_get_iff(struct net *net, struct file *file, struct ifreq *ifr)
{
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
struct tun_struct *tun = tun_get(file);
tun: TUNGETIFF interface to query name and flags Add a TUNGETIFF interface so that userspace can query a tun/tap descriptor for its name and flags. This is needed because it is common for one app to create a tap interface, exec another app and pass it the file descriptor for the interface. Without TUNGETIFF the spawned app has no way of detecting wheter the interface has e.g. IFF_VNET_HDR set. Signed-off-by: Mark McLoughlin <markmc@redhat.com> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-16 06:09:56 +08:00
if (!tun)
return -EBADFD;
DBG(KERN_INFO "%s: tun_get_iff\n", tun->dev->name);
strcpy(ifr->ifr_name, tun->dev->name);
ifr->ifr_flags = 0;
if (ifr->ifr_flags & TUN_TUN_DEV)
ifr->ifr_flags |= IFF_TUN;
else
ifr->ifr_flags |= IFF_TAP;
if (tun->flags & TUN_NO_PI)
ifr->ifr_flags |= IFF_NO_PI;
if (tun->flags & TUN_ONE_QUEUE)
ifr->ifr_flags |= IFF_ONE_QUEUE;
if (tun->flags & TUN_VNET_HDR)
ifr->ifr_flags |= IFF_VNET_HDR;
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
tun_put(tun);
tun: TUNGETIFF interface to query name and flags Add a TUNGETIFF interface so that userspace can query a tun/tap descriptor for its name and flags. This is needed because it is common for one app to create a tap interface, exec another app and pass it the file descriptor for the interface. Without TUNGETIFF the spawned app has no way of detecting wheter the interface has e.g. IFF_VNET_HDR set. Signed-off-by: Mark McLoughlin <markmc@redhat.com> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-16 06:09:56 +08:00
return 0;
}
tun: TUNSETFEATURES to set gso features. ethtool is useful for setting (some) device fields, but it's root-only. Finer feature control is available through a tun-specific ioctl. (Includes Mark McLoughlin <markmc@redhat.com>'s fix to hold rtnl sem). Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-03 18:46:16 +08:00
/* This is like a cut-down ethtool ops, except done via tun fd so no
* privs required. */
static int set_offload(struct net_device *dev, unsigned long arg)
{
unsigned int old_features, features;
old_features = dev->features;
/* Unset features, set them as we chew on the arg. */
features = (old_features & ~(NETIF_F_HW_CSUM|NETIF_F_SG|NETIF_F_FRAGLIST
|NETIF_F_TSO_ECN|NETIF_F_TSO|NETIF_F_TSO6));
if (arg & TUN_F_CSUM) {
features |= NETIF_F_HW_CSUM|NETIF_F_SG|NETIF_F_FRAGLIST;
arg &= ~TUN_F_CSUM;
if (arg & (TUN_F_TSO4|TUN_F_TSO6)) {
if (arg & TUN_F_TSO_ECN) {
features |= NETIF_F_TSO_ECN;
arg &= ~TUN_F_TSO_ECN;
}
if (arg & TUN_F_TSO4)
features |= NETIF_F_TSO;
if (arg & TUN_F_TSO6)
features |= NETIF_F_TSO6;
arg &= ~(TUN_F_TSO4|TUN_F_TSO6);
}
}
/* This gives the user a way to test for new features in future by
* trying to set them. */
if (arg)
return -EINVAL;
dev->features = features;
if (old_features != dev->features)
netdev_features_change(dev);
return 0;
}
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
static int tun_chr_ioctl(struct inode *inode, struct file *file,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
unsigned int cmd, unsigned long arg)
{
tun: Grab the netns in open. Grabbing namespaces in open, and putting them in close always seems to be the cleanest approach with the fewest surprises. So now that we have tun_file so we have somepleace to put the network namespace, let's grab the network namespace on file open and put on file close. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:01:48 +08:00
struct tun_file *tfile = file->private_data;
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
struct tun_struct *tun;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
void __user* argp = (void __user*)arg;
struct ifreq ifr;
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
int sndbuf;
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
int ret;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (cmd == TUNSETIFF || _IOC_TYPE(cmd) == 0x89)
if (copy_from_user(&ifr, argp, sizeof ifr))
return -EFAULT;
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
if (cmd == TUNGETFEATURES) {
/* Currently this just means: "what IFF flags are valid?".
* 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,
(unsigned int __user*)argp);
}
tun: Grab the netns in open. Grabbing namespaces in open, and putting them in close always seems to be the cleanest approach with the fewest surprises. So now that we have tun_file so we have somepleace to put the network namespace, let's grab the network namespace on file open and put on file close. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:01:48 +08:00
tun = __tun_get(tfile);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (cmd == TUNSETIFF && !tun) {
int err;
ifr.ifr_name[IFNAMSIZ-1] = '\0';
rtnl_lock();
tun: Grab the netns in open. Grabbing namespaces in open, and putting them in close always seems to be the cleanest approach with the fewest surprises. So now that we have tun_file so we have somepleace to put the network namespace, let's grab the network namespace on file open and put on file close. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:01:48 +08:00
err = tun_set_iff(tfile->net, file, &ifr);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
rtnl_unlock();
if (err)
return err;
if (copy_to_user(argp, &ifr, sizeof(ifr)))
return -EFAULT;
return 0;
}
tun: Interface to query tun/tap features. The problem with introducing checksum offload and gso to tun is they need to set dev->features to enable GSO and/or checksumming, which is supposed to be done before register_netdevice(), ie. as part of TUNSETIFF. Unfortunately, TUNSETIFF has always just ignored flags it doesn't understand, so there's no good way of detecting whether the kernel supports new IFF_ flags. This patch implements a TUNGETFEATURES ioctl which returns all the valid IFF flags. It could be extended later to include other features. Here's an example program which uses it: #include <linux/if_tun.h> #include <sys/types.h> #include <sys/ioctl.h> #include <sys/stat.h> #include <fcntl.h> #include <err.h> #include <stdio.h> static struct { unsigned int flag; const char *name; } known_flags[] = { { IFF_TUN, "TUN" }, { IFF_TAP, "TAP" }, { IFF_NO_PI, "NO_PI" }, { IFF_ONE_QUEUE, "ONE_QUEUE" }, }; int main() { unsigned int features, i; int netfd = open("/dev/net/tun", O_RDWR); if (netfd < 0) err(1, "Opening /dev/net/tun"); if (ioctl(netfd, TUNGETFEATURES, &features) != 0) { printf("Kernel does not support TUNGETFEATURES, guessing\n"); features = (IFF_TUN|IFF_TAP|IFF_NO_PI|IFF_ONE_QUEUE); } printf("Available features are: "); for (i = 0; i < sizeof(known_flags)/sizeof(known_flags[0]); i++) { if (features & known_flags[i].flag) { features &= ~known_flags[i].flag; printf("%s ", known_flags[i].name); } } if (features) printf("(UNKNOWN %#x)", features); printf("\n"); return 0; } Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-03 18:45:32 +08:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (!tun)
return -EBADFD;
DBG(KERN_INFO "%s: tun_chr_ioctl cmd %d\n", tun->dev->name, cmd);
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
ret = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
switch (cmd) {
tun: TUNGETIFF interface to query name and flags Add a TUNGETIFF interface so that userspace can query a tun/tap descriptor for its name and flags. This is needed because it is common for one app to create a tap interface, exec another app and pass it the file descriptor for the interface. Without TUNGETIFF the spawned app has no way of detecting wheter the interface has e.g. IFF_VNET_HDR set. Signed-off-by: Mark McLoughlin <markmc@redhat.com> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-16 06:09:56 +08:00
case TUNGETIFF:
ret = tun_get_iff(current->nsproxy->net_ns, file, &ifr);
if (ret)
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
break;
tun: TUNGETIFF interface to query name and flags Add a TUNGETIFF interface so that userspace can query a tun/tap descriptor for its name and flags. This is needed because it is common for one app to create a tap interface, exec another app and pass it the file descriptor for the interface. Without TUNGETIFF the spawned app has no way of detecting wheter the interface has e.g. IFF_VNET_HDR set. Signed-off-by: Mark McLoughlin <markmc@redhat.com> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-16 06:09:56 +08:00
if (copy_to_user(argp, &ifr, sizeof(ifr)))
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
ret = -EFAULT;
tun: TUNGETIFF interface to query name and flags Add a TUNGETIFF interface so that userspace can query a tun/tap descriptor for its name and flags. This is needed because it is common for one app to create a tap interface, exec another app and pass it the file descriptor for the interface. Without TUNGETIFF the spawned app has no way of detecting wheter the interface has e.g. IFF_VNET_HDR set. Signed-off-by: Mark McLoughlin <markmc@redhat.com> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-16 06:09:56 +08:00
break;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
case TUNSETNOCSUM:
/* Disable/Enable checksum */
if (arg)
tun->flags |= TUN_NOCHECKSUM;
else
tun->flags &= ~TUN_NOCHECKSUM;
DBG(KERN_INFO "%s: checksum %s\n",
tun->dev->name, arg ? "disabled" : "enabled");
break;
case TUNSETPERSIST:
/* Disable/Enable persist mode */
if (arg)
tun->flags |= TUN_PERSIST;
else
tun->flags &= ~TUN_PERSIST;
DBG(KERN_INFO "%s: persist %s\n",
[TUNTAP]: Fix wrong debug message. This is a trivial fix of debug message. When a persist flag is set, the message should say "enabled". Signed-off-by: Toyo Abe <tabe@miraclelinux.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-25 13:29:35 +08:00
tun->dev->name, arg ? "enabled" : "disabled");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
break;
case TUNSETOWNER:
/* Set owner of the device */
tun->owner = (uid_t) arg;
DBG(KERN_INFO "%s: owner set to %d\n", tun->dev->name, tun->owner);
break;
[NET]: Allow group ownership of TUN/TAP devices. Introduce a new syscall TUNSETGROUP for group ownership setting of tap devices. The user now is allowed to send packages if either his euid or his egid matches the one specified via tunctl (via -u or -g respecitvely). If both, gid and uid, are set via tunctl, both have to match. Signed-off-by: Guido Guenther <agx@sigxcpu.org> Signed-off-by: Jeff Dike <jdike@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-07-03 13:50:25 +08:00
case TUNSETGROUP:
/* Set group of the device */
tun->group= (gid_t) arg;
DBG(KERN_INFO "%s: group set to %d\n", tun->dev->name, tun->group);
break;
[TUNTAP]: Allow setting the linktype of the tap device from userspace Currently tun/tap only supports the EN10MB ARP type. For use with wireless and other networking types it should be possible to set the ARP type via an ioctl. Patch v2: Included check that the tap interface is down before changing the link type out from underneath it Signed-off-by: Mike Kershaw <dragorn@kismetwireless.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-09-02 08:40:05 +08:00
case TUNSETLINK:
/* Only allow setting the type when the interface is down */
tun: Fix minor race in TUNSETLINK ioctl handling. Noticed by Alan Cox. The IFF_UP test is a bit racey, because other entities outside of this driver's ioctl handler can modify that state, even though this ioctl handler runs under lock_kernel(). Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-24 10:37:58 +08:00
rtnl_lock();
[TUNTAP]: Allow setting the linktype of the tap device from userspace Currently tun/tap only supports the EN10MB ARP type. For use with wireless and other networking types it should be possible to set the ARP type via an ioctl. Patch v2: Included check that the tap interface is down before changing the link type out from underneath it Signed-off-by: Mike Kershaw <dragorn@kismetwireless.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-09-02 08:40:05 +08:00
if (tun->dev->flags & IFF_UP) {
DBG(KERN_INFO "%s: Linktype set failed because interface is up\n",
tun->dev->name);
tun: Fix minor race in TUNSETLINK ioctl handling. Noticed by Alan Cox. The IFF_UP test is a bit racey, because other entities outside of this driver's ioctl handler can modify that state, even though this ioctl handler runs under lock_kernel(). Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-24 10:37:58 +08:00
ret = -EBUSY;
[TUNTAP]: Allow setting the linktype of the tap device from userspace Currently tun/tap only supports the EN10MB ARP type. For use with wireless and other networking types it should be possible to set the ARP type via an ioctl. Patch v2: Included check that the tap interface is down before changing the link type out from underneath it Signed-off-by: Mike Kershaw <dragorn@kismetwireless.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-09-02 08:40:05 +08:00
} else {
tun->dev->type = (int) arg;
DBG(KERN_INFO "%s: linktype set to %d\n", tun->dev->name, tun->dev->type);
tun: Fix minor race in TUNSETLINK ioctl handling. Noticed by Alan Cox. The IFF_UP test is a bit racey, because other entities outside of this driver's ioctl handler can modify that state, even though this ioctl handler runs under lock_kernel(). Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-24 10:37:58 +08:00
ret = 0;
[TUNTAP]: Allow setting the linktype of the tap device from userspace Currently tun/tap only supports the EN10MB ARP type. For use with wireless and other networking types it should be possible to set the ARP type via an ioctl. Patch v2: Included check that the tap interface is down before changing the link type out from underneath it Signed-off-by: Mike Kershaw <dragorn@kismetwireless.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-09-02 08:40:05 +08:00
}
tun: Fix minor race in TUNSETLINK ioctl handling. Noticed by Alan Cox. The IFF_UP test is a bit racey, because other entities outside of this driver's ioctl handler can modify that state, even though this ioctl handler runs under lock_kernel(). Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-24 10:37:58 +08:00
rtnl_unlock();
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
break;
[TUNTAP]: Allow setting the linktype of the tap device from userspace Currently tun/tap only supports the EN10MB ARP type. For use with wireless and other networking types it should be possible to set the ARP type via an ioctl. Patch v2: Included check that the tap interface is down before changing the link type out from underneath it Signed-off-by: Mike Kershaw <dragorn@kismetwireless.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2005-09-02 08:40:05 +08:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
#ifdef TUN_DEBUG
case TUNSETDEBUG:
tun->debug = arg;
break;
#endif
tun: TUNSETFEATURES to set gso features. ethtool is useful for setting (some) device fields, but it's root-only. Finer feature control is available through a tun-specific ioctl. (Includes Mark McLoughlin <markmc@redhat.com>'s fix to hold rtnl sem). Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-03 18:46:16 +08:00
case TUNSETOFFLOAD:
rtnl_lock();
ret = set_offload(tun->dev, arg);
rtnl_unlock();
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
break;
tun: TUNSETFEATURES to set gso features. ethtool is useful for setting (some) device fields, but it's root-only. Finer feature control is available through a tun-specific ioctl. (Includes Mark McLoughlin <markmc@redhat.com>'s fix to hold rtnl sem). Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-03 18:46:16 +08:00
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
case TUNSETTXFILTER:
/* Can be set only for TAPs */
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
ret = -EINVAL;
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
if ((tun->flags & TUN_TYPE_MASK) != TUN_TAP_DEV)
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
break;
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
rtnl_lock();
net: tun.c fix cast Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-17 03:45:34 +08:00
ret = update_filter(&tun->txflt, (void __user *)arg);
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
rtnl_unlock();
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
break;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
case SIOCGIFHWADDR:
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
/* Get hw addres */
memcpy(ifr.ifr_hwaddr.sa_data, tun->dev->dev_addr, ETH_ALEN);
ifr.ifr_hwaddr.sa_family = tun->dev->type;
if (copy_to_user(argp, &ifr, sizeof ifr))
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
ret = -EFAULT;
break;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
case SIOCSIFHWADDR:
tun: Fix/rewrite packet filtering logic Please see the following thread to get some context on this http://marc.info/?l=linux-netdev&m=121564433018903&w=2 Basically the issue is that current multi-cast filtering stuff in the TUN/TAP driver is seriously broken. Original patch went in without proper review and ACK. It was broken and confusing to start with and subsequent patches broke it completely. To give you an idea of what's broken here are some of the issues: - Very confusing comments throughout the code that imply that the character device is a network interface in its own right, and that packets are passed between the two nics. Which is completely wrong. - Wrong set of ioctls is used for setting up filters. They look like shortcuts for manipulating state of the tun/tap network interface but in reality manipulate the state of the TX filter. - ioctls that were originally used for setting address of the the TX filter got "fixed" and now set the address of the network interface itself. Which made filter totaly useless. - Filtering is done too late. Instead of filtering early on, to avoid unnecessary wakeups, filtering is done in the read() call. The list goes on and on :) So the patch cleans all that up. It introduces simple and clean interface for setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering before enqueuing the packets. TX filtering is useful in the scenarios where TAP is part of a bridge, in which case it gets all broadcast, multicast and potentially other packets when the bridge is learning. So for example Ethernet tunnelling app may want to setup TX filters to avoid tunnelling multicast traffic. QEMU and other hypervisors can push RX filtering that is currently done in the guest into the host context therefore saving wakeups and unnecessary data transfer. Signed-off-by: Max Krasnyansky <maxk@qualcomm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-15 13:18:19 +08:00
/* Set hw address */
net: convert print_mac to %pM This converts pretty much everything to print_mac. There were a few things that had conflicts which I have just dropped for now, no harm done. I've built an allyesconfig with this and looked at the files that weren't built very carefully, but it's a huge patch. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-28 06:59:26 +08:00
DBG(KERN_DEBUG "%s: set hw address: %pM\n",
tun->dev->name, ifr.ifr_hwaddr.sa_data);
[TUN]: Fix RTNL-locking in tun/tap driver Current tun/tap driver sets also net device's hw address when asked to change character device's hw address. This is a good idea, but it misses RTLN-locking, resulting following error message in 2.6.25-rc3's inetdev_event() function: RTNL: assertion failed at net/ipv4/devinet.c (1050) Attached patch fixes this problem. Signed-off-by: Kim B. Heino <Kim.Heino@bluegiga.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-03-01 04:26:21 +08:00
rtnl_lock();
ret = dev_set_mac_address(tun->dev, &ifr.ifr_hwaddr);
rtnl_unlock();
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
break;
tun: Limit amount of queued packets per device Unlike a normal socket path, the tuntap device send path does not have any accounting. This means that the user-space sender may be able to pin down arbitrary amounts of kernel memory by continuing to send data to an end-point that is congested. Even when this isn't an issue because of limited queueing at most end points, this can also be a problem because its only response to congestion is packet loss. That is, when those local queues at the end-point fills up, the tuntap device will start wasting system time because it will continue to send data there which simply gets dropped straight away. Of course one could argue that everybody should do congestion control end-to-end, unfortunately there are people in this world still hooked on UDP, and they don't appear to be going away anywhere fast. In fact, we've always helped them by performing accounting in our UDP code, the sole purpose of which is to provide congestion feedback other than through packet loss. This patch attempts to apply the same bandaid to the tuntap device. It creates a pseudo-socket object which is used to account our packets just as a normal socket does for UDP. Of course things are a little complex because we're actually reinjecting traffic back into the stack rather than out of the stack. The stack complexities however should have been resolved by preceding patches. So this one can simply start using skb_set_owner_w. For now the accounting is essentially disabled by default for backwards compatibility. In particular, we set the cap to INT_MAX. This is so that existing applications don't get confused by the sudden arrival EAGAIN errors. In future we may wish (or be forced to) do this by default. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 13:25:32 +08:00
case TUNGETSNDBUF:
sndbuf = tun->sk->sk_sndbuf;
if (copy_to_user(argp, &sndbuf, sizeof(sndbuf)))
ret = -EFAULT;
break;
case TUNSETSNDBUF:
if (copy_from_user(&sndbuf, argp, sizeof(sndbuf))) {
ret = -EFAULT;
break;
}
tun->sk->sk_sndbuf = sndbuf;
break;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
default:
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
ret = -EINVAL;
break;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
};
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
tun_put(tun);
return ret;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
static int tun_chr_fasync(int fd, struct file *file, int on)
{
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
struct tun_struct *tun = tun_get(file);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
int ret;
if (!tun)
return -EBADFD;
DBG(KERN_INFO "%s: tun_chr_fasync %d\n", tun->dev->name, on);
tun: fasync BKL pushdown Signed-off-by: Jonathan Corbet <corbet@lwn.net>
2008-06-20 05:50:37 +08:00
lock_kernel();
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if ((ret = fasync_helper(fd, file, on, &tun->fasync)) < 0)
tun: fasync BKL pushdown Signed-off-by: Jonathan Corbet <corbet@lwn.net>
2008-06-20 05:50:37 +08:00
goto out;
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (on) {
[PATCH] file: modify struct fown_struct to use a struct pid File handles can be requested to send sigio and sigurg to processes. By tracking the destination processes using struct pid instead of pid_t we make the interface safe from all potential pid wrap around problems. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-02 17:17:15 +08:00
ret = __f_setown(file, task_pid(current), PIDTYPE_PID, 0);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (ret)
tun: fasync BKL pushdown Signed-off-by: Jonathan Corbet <corbet@lwn.net>
2008-06-20 05:50:37 +08:00
goto out;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
tun->flags |= TUN_FASYNC;
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
} else
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
tun->flags &= ~TUN_FASYNC;
tun: fasync BKL pushdown Signed-off-by: Jonathan Corbet <corbet@lwn.net>
2008-06-20 05:50:37 +08:00
ret = 0;
out:
unlock_kernel();
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
tun_put(tun);
tun: fasync BKL pushdown Signed-off-by: Jonathan Corbet <corbet@lwn.net>
2008-06-20 05:50:37 +08:00
return ret;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
static int tun_chr_open(struct inode *inode, struct file * file)
{
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
struct tun_file *tfile;
net-tun: BKL pushdown Signed-off-by: Arnd Bergmann <arnd@arndb.de>
2008-05-21 01:16:24 +08:00
cycle_kernel_lock();
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
DBG1(KERN_INFO "tunX: tun_chr_open\n");
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
tfile = kmalloc(sizeof(*tfile), GFP_KERNEL);
if (!tfile)
return -ENOMEM;
tun: Fix races between tun_net_close and free_netdev. The tun code does not cope gracefully if the network device goes away before the tun file descriptor is closed. It looks like we can trigger this with rmmod, and moving tun devices between network namespaces will allow this to be triggered when network namespaces exit. To fix this I introduce an intermediate data structure tun_file which holds a count of users and a pointer to the struct tun_struct. tun_get increments that reference count if it is greater than 0. tun_put decrements that reference count and detaches from the network device if the count is 0. While we have a file attached to the network device I hold a reference to the network device keeping it from going away completely. When a network device is unregistered I decrement the count of the attached tun_file and if that was the last user I detach the tun_file, and all processes on read_wait are woken up to ensure they do not sleep indefinitely. As some of those sleeps happen with the count on the tun device elevated waking up the read waiters ensures that tun_file will be detached in a timely manner. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:07:17 +08:00
atomic_set(&tfile->count, 0);
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
tfile->tun = NULL;
tun: Grab the netns in open. Grabbing namespaces in open, and putting them in close always seems to be the cleanest approach with the fewest surprises. So now that we have tun_file so we have somepleace to put the network namespace, let's grab the network namespace on file open and put on file close. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:01:48 +08:00
tfile->net = get_net(current->nsproxy->net_ns);
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
file->private_data = tfile;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return 0;
}
static int tun_chr_close(struct inode *inode, struct file *file)
{
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
struct tun_file *tfile = file->private_data;
struct tun_struct *tun = __tun_get(tfile);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
if (tun) {
DBG(KERN_INFO "%s: tun_chr_close\n", tun->dev->name);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
rtnl_lock();
__tun_detach(tun);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
/* If desireable, unregister the netdevice. */
tun: Only free a netdev when all tun descriptors are closed The commit c70f182940f988448f3c12a209d18b1edc276e33 ("tun: Fix races between tun_net_close and free_netdev") fixed a race where an asynchronous deletion of a tun device can hose a poll(2) on a tun fd attached to that device. However, this came at the cost of moving the tun wait queue into the tun file data structure. The problem with this is that it imposes restrictions on when and where the tun device can access the wait queue since the tun file may change at any time due to detaching and reattaching. In particular, now that we need to use the wait queue on the receive path it becomes difficult to properly synchronise this with the detachment of the tun device. This patch solves the original race in a different way. Since the race is only because the underlying memory gets freed, we can prevent it simply by ensuring that we don't do that until all tun descriptors ever attached to the device (even if they have since be detached because they may still be sitting in poll) have been closed. This is done by using reference counting the attached tun file descriptors. The refcount in tun->sk has been reappropriated for this purpose since it was already being used for that, albeit from the opposite angle. Note that we no longer zero tfile->tun since tun_get will return NULL anyway after the refcount on tfile hits zero. Instead it represents whether this device has ever been attached to a device. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-18 22:15:52 +08:00
if (!(tun->flags & TUN_PERSIST))
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
unregister_netdevice(tun->dev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
rtnl_unlock();
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
tun: Only free a netdev when all tun descriptors are closed The commit c70f182940f988448f3c12a209d18b1edc276e33 ("tun: Fix races between tun_net_close and free_netdev") fixed a race where an asynchronous deletion of a tun device can hose a poll(2) on a tun fd attached to that device. However, this came at the cost of moving the tun wait queue into the tun file data structure. The problem with this is that it imposes restrictions on when and where the tun device can access the wait queue since the tun file may change at any time due to detaching and reattaching. In particular, now that we need to use the wait queue on the receive path it becomes difficult to properly synchronise this with the detachment of the tun device. This patch solves the original race in a different way. Since the race is only because the underlying memory gets freed, we can prevent it simply by ensuring that we don't do that until all tun descriptors ever attached to the device (even if they have since be detached because they may still be sitting in poll) have been closed. This is done by using reference counting the attached tun file descriptors. The refcount in tun->sk has been reappropriated for this purpose since it was already being used for that, albeit from the opposite angle. Note that we no longer zero tfile->tun since tun_get will return NULL anyway after the refcount on tfile hits zero. Instead it represents whether this device has ever been attached to a device. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-18 22:15:52 +08:00
tun = tfile->tun;
if (tun)
sock_put(tun->sk);
tun: Grab the netns in open. Grabbing namespaces in open, and putting them in close always seems to be the cleanest approach with the fewest surprises. So now that we have tun_file so we have somepleace to put the network namespace, let's grab the network namespace on file open and put on file close. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:01:48 +08:00
put_net(tfile->net);
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
kfree(tfile);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return 0;
}
[PATCH] mark struct file_operations const 5 Many struct file_operations in the kernel can be "const". Marking them const moves these to the .rodata section, which avoids false sharing with potential dirty data. In addition it'll catch accidental writes at compile time to these shared resources. Signed-off-by: Arjan van de Ven <arjan@linux.intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-12 16:55:34 +08:00
static const struct file_operations tun_fops = {
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
.owner = THIS_MODULE,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
.llseek = no_llseek,
[PATCH] Remove readv/writev methods and use aio_read/aio_write instead This patch removes readv() and writev() methods and replaces them with aio_read()/aio_write() methods. Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-01 14:28:47 +08:00
.read = do_sync_read,
.aio_read = tun_chr_aio_read,
.write = do_sync_write,
.aio_write = tun_chr_aio_write,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
.poll = tun_chr_poll,
.ioctl = tun_chr_ioctl,
.open = tun_chr_open,
.release = tun_chr_close,
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
.fasync = tun_chr_fasync
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
};
static struct miscdevice tun_miscdev = {
.minor = TUN_MINOR,
.name = "tun",
.fops = &tun_fops,
};
/* ethtool interface */
static int tun_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
cmd->supported = 0;
cmd->advertising = 0;
cmd->speed = SPEED_10;
cmd->duplex = DUPLEX_FULL;
cmd->port = PORT_TP;
cmd->phy_address = 0;
cmd->transceiver = XCVR_INTERNAL;
cmd->autoneg = AUTONEG_DISABLE;
cmd->maxtxpkt = 0;
cmd->maxrxpkt = 0;
return 0;
}
static void tun_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
{
struct tun_struct *tun = netdev_priv(dev);
strcpy(info->driver, DRV_NAME);
strcpy(info->version, DRV_VERSION);
strcpy(info->fw_version, "N/A");
switch (tun->flags & TUN_TYPE_MASK) {
case TUN_TUN_DEV:
strcpy(info->bus_info, "tun");
break;
case TUN_TAP_DEV:
strcpy(info->bus_info, "tap");
break;
}
}
static u32 tun_get_msglevel(struct net_device *dev)
{
#ifdef TUN_DEBUG
struct tun_struct *tun = netdev_priv(dev);
return tun->debug;
#else
return -EOPNOTSUPP;
#endif
}
static void tun_set_msglevel(struct net_device *dev, u32 value)
{
#ifdef TUN_DEBUG
struct tun_struct *tun = netdev_priv(dev);
tun->debug = value;
#endif
}
static u32 tun_get_link(struct net_device *dev)
{
struct tun_struct *tun = netdev_priv(dev);
tun: Introduce tun_file Currently the tun code suffers from only having a single word of data that exists for the entire life of the tun file descriptor. This results in peculiar holding of references to the network namespace as well as races between free_netdevice and tun_chr_close. Fix this by introducing tun_file which will hold the per file state. For the moment it still holds just a single word so the differences are all logic changes with no changes in semantics. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-20 19:00:40 +08:00
return !!tun->tfile;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
static u32 tun_get_rx_csum(struct net_device *dev)
{
struct tun_struct *tun = netdev_priv(dev);
return (tun->flags & TUN_NOCHECKSUM) == 0;
}
static int tun_set_rx_csum(struct net_device *dev, u32 data)
{
struct tun_struct *tun = netdev_priv(dev);
if (data)
tun->flags &= ~TUN_NOCHECKSUM;
else
tun->flags |= TUN_NOCHECKSUM;
return 0;
}
drivers/net: const-ify ethtool_ops declarations Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 02:30:00 +08:00
static const struct ethtool_ops tun_ethtool_ops = {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
.get_settings = tun_get_settings,
.get_drvinfo = tun_get_drvinfo,
.get_msglevel = tun_get_msglevel,
.set_msglevel = tun_set_msglevel,
.get_link = tun_get_link,
.get_rx_csum = tun_get_rx_csum,
.set_rx_csum = tun_set_rx_csum
};
[TUN]: Introduce the tun_net structure and init/exit net ops. This is the first step in making tuntap devices work in net namespaces. The structure mentioned is pointed by generic net pointer with tun_net_id id, and tun driver fills one on its load. It will contain only the tun devices list. So declare this structure and introduce net init and exit hooks. Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-16 15:40:46 +08:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
static int __init tun_init(void)
{
int ret = 0;
printk(KERN_INFO "tun: %s, %s\n", DRV_DESCRIPTION, DRV_VERSION);
printk(KERN_INFO "tun: %s\n", DRV_COPYRIGHT);
tun: Implement ip link del tunXXX This greatly simplifies testing to verify I have fixed the problems with a tun device disappearing when the tun file descriptor is still held open. Further it allows removal network namespace operations for the tun driver. Reducing the network namespace handling in the driver to the minimum. i.e. When we are creating a tun device. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-22 08:02:16 +08:00
ret = rtnl_link_register(&tun_link_ops);
[TUN]: Introduce the tun_net structure and init/exit net ops. This is the first step in making tuntap devices work in net namespaces. The structure mentioned is pointed by generic net pointer with tun_net_id id, and tun driver fills one on its load. It will contain only the tun devices list. So declare this structure and introduce net init and exit hooks. Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-16 15:40:46 +08:00
if (ret) {
tun: Implement ip link del tunXXX This greatly simplifies testing to verify I have fixed the problems with a tun device disappearing when the tun file descriptor is still held open. Further it allows removal network namespace operations for the tun driver. Reducing the network namespace handling in the driver to the minimum. i.e. When we are creating a tun device. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-22 08:02:16 +08:00
printk(KERN_ERR "tun: Can't register link_ops\n");
goto err_linkops;
[TUN]: Introduce the tun_net structure and init/exit net ops. This is the first step in making tuntap devices work in net namespaces. The structure mentioned is pointed by generic net pointer with tun_net_id id, and tun driver fills one on its load. It will contain only the tun devices list. So declare this structure and introduce net init and exit hooks. Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-16 15:40:46 +08:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
ret = misc_register(&tun_miscdev);
[TUN]: Introduce the tun_net structure and init/exit net ops. This is the first step in making tuntap devices work in net namespaces. The structure mentioned is pointed by generic net pointer with tun_net_id id, and tun driver fills one on its load. It will contain only the tun devices list. So declare this structure and introduce net init and exit hooks. Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-16 15:40:46 +08:00
if (ret) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
printk(KERN_ERR "tun: Can't register misc device %d\n", TUN_MINOR);
[TUN]: Introduce the tun_net structure and init/exit net ops. This is the first step in making tuntap devices work in net namespaces. The structure mentioned is pointed by generic net pointer with tun_net_id id, and tun driver fills one on its load. It will contain only the tun devices list. So declare this structure and introduce net init and exit hooks. Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-16 15:40:46 +08:00
goto err_misc;
}
tun: Implement ip link del tunXXX This greatly simplifies testing to verify I have fixed the problems with a tun device disappearing when the tun file descriptor is still held open. Further it allows removal network namespace operations for the tun driver. Reducing the network namespace handling in the driver to the minimum. i.e. When we are creating a tun device. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-22 08:02:16 +08:00
return 0;
[TUN]: Introduce the tun_net structure and init/exit net ops. This is the first step in making tuntap devices work in net namespaces. The structure mentioned is pointed by generic net pointer with tun_net_id id, and tun driver fills one on its load. It will contain only the tun devices list. So declare this structure and introduce net init and exit hooks. Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-16 15:40:46 +08:00
err_misc:
tun: Implement ip link del tunXXX This greatly simplifies testing to verify I have fixed the problems with a tun device disappearing when the tun file descriptor is still held open. Further it allows removal network namespace operations for the tun driver. Reducing the network namespace handling in the driver to the minimum. i.e. When we are creating a tun device. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-22 08:02:16 +08:00
rtnl_link_unregister(&tun_link_ops);
err_linkops:
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return ret;
}
static void tun_cleanup(void)
{
drivers/net: Trim trailing whitespace Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-14 01:24:59 +08:00
misc_deregister(&tun_miscdev);
tun: Implement ip link del tunXXX This greatly simplifies testing to verify I have fixed the problems with a tun device disappearing when the tun file descriptor is still held open. Further it allows removal network namespace operations for the tun driver. Reducing the network namespace handling in the driver to the minimum. i.e. When we are creating a tun device. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-22 08:02:16 +08:00
rtnl_link_unregister(&tun_link_ops);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
module_init(tun_init);
module_exit(tun_cleanup);
MODULE_DESCRIPTION(DRV_DESCRIPTION);
MODULE_AUTHOR(DRV_COPYRIGHT);
MODULE_LICENSE("GPL");
MODULE_ALIAS_MISCDEV(TUN_MINOR);
Copy Permalink