Skip to content

L
linux
Commit 4ce937cb authored by Linus Torvalds's avatar Linus Torvalds
Browse files
Options

Merge master.kernel.org:/home/davem/BK/net-2.5 

into home.transmeta.com:/home/torvalds/v2.5/linux
parents 80d504c8 ff5d22a4
Hide whitespace changes
Inline Side-by-side
Showing with 2993 additions and 150 deletions
Documentation/networking/NAPI_HOWTO.txt 0 → 100644
View file @ 4ce937cb
HISTORY:
February 16/2002 -- revision 0.2.1:
COR typo corrected
February 10/2002 -- revision 0.2:
some spell checking ;->
January 12/2002 -- revision 0.1
This is still work in progress so may change.
To keep up to date please watch this space.
Introduction to NAPI
====================
NAPI is a proven (www.cyberus.ca/~hadi/usenix-paper.tgz) technique
to improve network performance on Linux. For more details please
read that paper.
NAPI provides a "inherent mitigation" which is bound by system capacity
as can be seen from the following data collected by Robert on Gigabit
ethernet (e1000):
Psize Ipps Tput Rxint Txint Done Ndone
---------------------------------------------------------------
60 890000 409362 17 27622 7 6823
128 758150 464364 21 9301 10 7738
256 445632 774646 42 15507 21 12906
512 232666 994445 241292 19147 241192 1062
1024 119061 1000003 872519 19258 872511 0
1440 85193 1000003 946576 19505 946569 0
Legend:
"Ipps" stands for input packets per second.
"Tput" == packets out of total 1M that made it out.
"txint" == transmit completion interrupts seen
"Done" == The number of times that the poll() managed to pull all
packets out of the rx ring. Note from this that the lower the
load the more we could clean up the rxring
"Ndone" == is the converse of "Done". Note again, that the higher
the load the more times we couldnt clean up the rxring.
Observe that:
when the NIC receives 890Kpackets/sec only 17 rx interrupts are generated.
The system cant handle the processing at 1 interrupt/packet at that load level.
At lower rates on the other hand, rx interrupts go up and therefore the
interrupt/packet ratio goes up (as observable from that table). So there is
possibility that under low enough input, you get one poll call for each
input packet caused by a single interrupt each time. And if the system
cant handle interrupt per packet ratio of 1, then it will just have to
chug along ....
0) Prerequisites:
==================
A driver MAY continue using the old 2.4 technique for interfacing
to the network stack and not benefit from the NAPI changes.
NAPI additions to the kernel do not break backward compatibility.
NAPI, however, requires the following features to be available:
A) DMA ring or enough RAM to store packets in software devices.
B) Ability to turn off interrupts or maybe events that send packets up
the stack.
NAPI processes packet events in what is known as dev->poll() method.
Typically, only packet receive events are processed in dev->poll().
The rest of the events MAY be processed by the regular interrupt handler
to reduce processing latency (justified also because there are not that
many of them).
Note, however, NAPI does not enforce that dev->poll() only processes
receive events.
Tests with the tulip driver indicated slightly increased latency if
all of the interrupt handler is moved to dev->poll(). Also MII handling
gets a little trickier.
The example used in this document is to move the receive processing only
to dev->poll(); this is shown with the patch for the tulip driver.
For an example of code that moves all the interrupt driver to
dev->poll() look at the ported e1000 code.
There are caveats that might force you to go with moving everything to
dev->poll(). Different NICs work differently depending on their status/event
acknowledgement setup.
There are two types of event register ACK mechanisms.
I) what is known as Clear-on-read (COR).
when you read the status/event register, it clears everything!
The natsemi and sunbmac NICs are known to do this.
In this case your only choice is to move all to dev->poll()
II) Clear-on-write (COW)
i) you clear the status by writting a 1 in the bit-location you want.
These are the majority of the NICs and work the best with NAPI.
Put only receive events in dev->poll(); leave the rest in
the old interrupt handler.
ii) whatever you write in the status register clears every thing ;->
Cant seem to find any supported by Linux which do this. If
someone knows such a chip email us please.
Move all to dev->poll()
C) Ability to detect new work correctly.
NAPI works by shutting down event interrupts when theres work and
turning them on when theres none.
New packets might show up in the small window while interrupts were being
re-enabled (refer to appendix 2). A packet might sneak in during the period
we are enabling interrupts. We only get to know about such a packet when the
next new packet arrives and generates an interrupt.
Essentially, there is a small window of opportunity for a race condition
which for clarity we'll refer to as the "rotting packet".
This is a very important topic and appendix 2 is dedicated for more
discussion.
Locking rules and environmental guarantees
==========================================
-Guarantee: Only one CPU at any time can call dev->poll(); this is because
only one CPU can pick the initial interrupt and hence the initial
netif_rx_schedule(dev);
- The core layer invokes devices to send packets in a round robin format.
This implies receive is totaly lockless because of the guarantee only that
one CPU is executing it.
- contention can only be the result of some other CPU accessing the rx
ring. This happens only in close() and suspend() (when these methods
try to clean the rx ring);
****guarantee: driver authors need not worry about this; synchronization
is taken care for them by the top net layer.
-local interrupts are enabled (if you dont move all to dev->poll()). For
example link/MII and txcomplete continue functioning just same old way.
This improves the latency of processing these events. It is also assumed that
the receive interrupt is the largest cause of noise. Note this might not
always be true.
[according to Manfred Spraul, the winbond insists on sending one
txmitcomplete interrupt for each packet (although this can be mitigated)].
For these broken drivers, move all to dev->poll().
For the rest of this text, we'll assume that dev->poll() only
processes receive events.
new methods introduce by NAPI
=============================
a) netif_rx_schedule(dev)
Called by an IRQ handler to schedule a poll for device
b) netif_rx_schedule_prep(dev)
puts the device in a state which allows for it to be added to the
CPU polling list if it is up and running. You can look at this as
the first half of netif_rx_schedule(dev) above; the second half
being c) below.
c) __netif_rx_schedule(dev)
Add device to the poll list for this CPU; assuming that _prep above
has already been called and returned 1.
d) netif_rx_reschedule(dev, undo)
Called to reschedule polling for device specifically for some
deficient hardware. Read Appendix 2 for more details.
e) netif_rx_complete(dev)
Remove interface from the CPU poll list: it must be in the poll list
on current cpu. This primitive is called by dev->poll(), when
it completes its work. The device cannot be out of poll list at this
call, if it is then clearly it is a BUG(). You'll know ;->
All these above nethods are used below. So keep reading for clarity.
Device driver changes to be made when porting NAPI
==================================================
Below we describe what kind of changes are required for NAPI to work.
1) introduction of dev->poll() method
=====================================
This is the method that is invoked by the network core when it requests
for new packets from the driver. A driver is allowed to send upto
dev->quota packets by the current CPU before yielding to the network
subsystem (so other devices can also get opportunity to send to the stack).
dev->poll() prototype looks as follows:
int my_poll(struct net_device *dev, int *budget)
budget is the remaining number of packets the network subsystem on the
current CPU can send up the stack before yielding to other system tasks.
*Each driver is responsible for decrementing budget by the total number of
packets sent.
Total number of packets cannot exceed dev->quota.
dev->poll() method is invoked by the top layer, the driver just sends if it
can to the stack the packet quantity requested.
more on dev->poll() below after the interrupt changes are explained.
2) registering dev->poll() method
===================================
dev->poll should be set in the dev->probe() method.
e.g:
dev->open = my_open;
.
.
/* two new additions */
/* first register my poll method */
dev->poll = my_poll;
/* next register my weight/quanta; can be overriden in /proc */
dev->weight = 16;
.
.
dev->stop = my_close;
3) scheduling dev->poll()
=============================
This involves modifying the interrupt handler and the code
path which takes the packet off the NIC and sends them to the
stack.
it's important at this point to introduce the classical D Becker
interrupt processor:
------------------
static void
netdevice_interrupt(int irq, void *dev_id, struct pt_regs *regs)
{
struct net_device *dev = (struct net_device *)dev_instance;
struct my_private *tp = (struct my_private *)dev->priv;
int work_count = my_work_count;
status = read_interrupt_status_reg();
if (status == 0)
return; /* Shared IRQ: not us */
if (status == 0xffff)
return; /* Hot unplug */
if (status & error)
do_some_error_handling()
do {
acknowledge_ints_ASAP();
if (status & link_interrupt) {
spin_lock(&tp->link_lock);
do_some_link_stat_stuff();
spin_lock(&tp->link_lock);
}
if (status & rx_interrupt) {
receive_packets(dev);
}
if (status & rx_nobufs) {
make_rx_buffs_avail();
}
if (status & tx_related) {
spin_lock(&tp->lock);
tx_ring_free(dev);
if (tx_died)
restart_tx();
spin_unlock(&tp->lock);
}
status = read_interrupt_status_reg();
} while (!(status & error) || more_work_to_be_done);
}
----------------------------------------------------------------------
We now change this to what is shown below to NAPI-enable it:
----------------------------------------------------------------------
static void
netdevice_interrupt(int irq, void *dev_id, struct pt_regs *regs)
{
struct net_device *dev = (struct net_device *)dev_instance;
struct my_private *tp = (struct my_private *)dev->priv;
status = read_interrupt_status_reg();
if (status == 0)
return; /* Shared IRQ: not us */
if (status == 0xffff)
return; /* Hot unplug */
if (status & error)
do_some_error_handling();
do {
/************************ start note *********************************/
acknowledge_ints_ASAP(); // dont ack rx and rxnobuff here
/************************ end note *********************************/
if (status & link_interrupt) {
spin_lock(&tp->link_lock);
do_some_link_stat_stuff();
spin_unlock(&tp->link_lock);
}
/************************ start note *********************************/
if (status & rx_interrupt || (status & rx_nobuffs)) {
if (netif_rx_schedule_prep(dev)) {
/* disable interrupts caused
* by arriving packets */
disable_rx_and_rxnobuff_ints();
/* tell system we have work to be done. */
__netif_rx_schedule(dev);
} else {
printk("driver bug! interrupt while in poll\n");
/* FIX by disabling interrupts */
disable_rx_and_rxnobuff_ints();
}
}
/************************ end note note *********************************/
if (status & tx_related) {
spin_lock(&tp->lock);
tx_ring_free(dev);
if (tx_died)
restart_tx();
spin_unlock(&tp->lock);
}
status = read_interrupt_status_reg();
/************************ start note *********************************/
} while (!(status & error) || more_work_to_be_done(status));
/************************ end note note *********************************/
}
---------------------------------------------------------------------
We note several things from above:
I) Any interrupt source which is caused by arriving packets is now
turned off when it occurs. Depending on the hardware, there could be
several reasons that arriving packets would cause interrupts; these are the
interrupt sources we wish to avoid. The two common ones are a) a packet
arriving (rxint) b) a packet arriving and finding no DMA buffers available
(rxnobuff) .
This means also acknowledge_ints_ASAP() will not clear the status
register for those two items above; clearing is done in the place where
proper work is done within NAPI; at the poll() and refill_rx_ring()
discussed further below.
netif_rx_schedule_prep() returns 1 if device is in running state and
gets successfully added to the core poll list. If we get a zero value
we can _almost_ assume are already added to the list (instead of not running.
Logic based on the fact that you shouldnt get interrupt if not running)
We rectify this by disabling rx and rxnobuf interrupts.
II) that receive_packets(dev) and make_rx_buffs_avail() may have dissapeared.
These functionalities are still around actually......
infact, receive_packets(dev) is very close to my_poll() and
make_rx_buffs_avail() is invoked from my_poll()
4) converting receive_packets() to dev->poll()
===============================================
We need to convert the classical D Becker receive_packets(dev) to my_poll()
First the typical receive_packets() below:
-------------------------------------------------------------------
/* this is called by interrupt handler */
static void receive_packets (struct net_device *dev)
{
struct my_private *tp = (struct my_private *)dev->priv;
rx_ring = tp->rx_ring;
cur_rx = tp->cur_rx;
int entry = cur_rx % RX_RING_SIZE;
int received = 0;
int rx_work_limit = tp->dirty_rx + RX_RING_SIZE - tp->cur_rx;
while (rx_ring_not_empty) {
u32 rx_status;
unsigned int rx_size;
unsigned int pkt_size;
struct sk_buff *skb;
/* read size+status of next frame from DMA ring buffer */
/* the number 16 and 4 are just examples */
rx_status = le32_to_cpu (*(u32 *) (rx_ring + ring_offset));
rx_size = rx_status >> 16;
pkt_size = rx_size - 4;
/* process errors */
if ((rx_size > (MAX_ETH_FRAME_SIZE+4)) ||
(!(rx_status & RxStatusOK))) {
netdrv_rx_err (rx_status, dev, tp, ioaddr);
return;
}
if (--rx_work_limit < 0)
break;
/* grab a skb */
skb = dev_alloc_skb (pkt_size + 2);
if (skb) {
.
.
netif_rx (skb);
.
.
} else { /* OOM */
/*seems very driver specific ... some just pass
whatever is on the ring already. */
}
/* move to the next skb on the ring */
entry = (++tp->cur_rx) % RX_RING_SIZE;
received++ ;
}
/* store current ring pointer state */
tp->cur_rx = cur_rx;
/* Refill the Rx ring buffers if they are needed */
refill_rx_ring();
.
.
}
-------------------------------------------------------------------
We change it to a new one below; note the additional parameter in
the call.
-------------------------------------------------------------------
/* this is called by the network core */
static void my_poll (struct net_device *dev, int *budget)
{
struct my_private *tp = (struct my_private *)dev->priv;
rx_ring = tp->rx_ring;
cur_rx = tp->cur_rx;
int entry = cur_rx % RX_BUF_LEN;
/* maximum packets to send to the stack */
/************************ note note *********************************/
int rx_work_limit = dev->quota;
/************************ end note note *********************************/
do { // outer beggining loop starts here
clear_rx_status_register_bit();
while (rx_ring_not_empty) {
u32 rx_status;
unsigned int rx_size;
unsigned int pkt_size;
struct sk_buff *skb;
/* read size+status of next frame from DMA ring buffer */
/* the number 16 and 4 are just examples */
rx_status = le32_to_cpu (*(u32 *) (rx_ring + ring_offset));
rx_size = rx_status >> 16;
pkt_size = rx_size - 4;
/* process errors */
if ((rx_size > (MAX_ETH_FRAME_SIZE+4)) ||
(!(rx_status & RxStatusOK))) {
netdrv_rx_err (rx_status, dev, tp, ioaddr);
return;
}
/************************ note note *********************************/
if (--rx_work_limit < 0) { /* we got packets, but no quota */
/* store current ring pointer state */
tp->cur_rx = cur_rx;
/* Refill the Rx ring buffers if they are needed */
refill_rx_ring(dev);
goto not_done;
}
/********************** end note **********************************/
/* grab a skb */
skb = dev_alloc_skb (pkt_size + 2);
if (skb) {
.
.
/************************ note note *********************************/
netif_receive_skb (skb);
/********************** end note **********************************/
.
.
} else { /* OOM */
/*seems very driver specific ... common is just pass
whatever is on the ring already. */
}
/* move to the next skb on the ring */
entry = (++tp->cur_rx) % RX_RING_SIZE;
received++ ;
}
/* store current ring pointer state */
tp->cur_rx = cur_rx;
/* Refill the Rx ring buffers if they are needed */
refill_rx_ring(dev);
/* no packets on ring; but new ones can arrive since we last
checked */
status = read_interrupt_status_reg();
if (rx status is not set) {
/* If something arrives in this narrow window,
an interrupt will be generated */
goto done;
}
/* done! at least thats what it looks like ;->
if new packets came in after our last check on status bits
they'll be caught by the while check and we go back and clear them
since we havent exceeded our quota */
} while (rx_status_is_set);
done:
/************************ note note *********************************/
dev->quota -= received;
*budget -= received;
/* If RX ring is not full we are out of memory. */
if (tp->rx_buffers[tp->dirty_rx % RX_RING_SIZE].skb == NULL)
goto oom;
/* we are happy/done, no more packets on ring; put us back
to where we can start processing interrupts again */
netif_rx_complete(dev);
enable_rx_and_rxnobuf_ints();
/* The last op happens after poll completion. Which means the following:
* 1. it can race with disabling irqs in irq handler (which are done to
* schedule polls)
* 2. it can race with dis/enabling irqs in other poll threads
* 3. if an irq raised after the begining of the outer beginning
* loop(marked in the code above), it will be immediately
* triggered here.
*
* Summarizing: the logic may results in some redundant irqs both
* due to races in masking and due to too late acking of already
* processed irqs. The good news: no events are ever lost.
*/
return 0; /* done */
not_done:
if (tp->cur_rx - tp->dirty_rx > RX_RING_SIZE/2 ||
tp->rx_buffers[tp->dirty_rx % RX_RING_SIZE].skb == NULL)
refill_rx_ring(dev);
if (!received) {
printk("received==0\n");
received = 1;
}
dev->quota -= received;
*budget -= received;
return 1; /* not_done */
oom:
/* Start timer, stop polling, but do not enable rx interrupts. */
start_poll_timer(dev);
return 0; /* we'll take it from here so tell core "done"*/
/************************ End note note *********************************/
}
-------------------------------------------------------------------
From above we note that:
0) rx_work_limit = dev->quota
1) refill_rx_ring() is in charge of clearing the bit for rxnobuff when
it does the work.
2) We have a done and not_done state.
3) instead of netif_rx() we call netif_receive_skb() to pass the skb.
4) we have a new way of handling oom condition
5) A new outer for (;;) loop has been added. This serves the purpose of
ensuring that if a new packet has come in, after we are all set and done,
and we have not exceeded our quota that we continue sending packets up.
-----------------------------------------------------------
Poll timer code will need to do the following:
a)
if (tp->cur_rx - tp->dirty_rx > RX_RING_SIZE/2 ||
tp->rx_buffers[tp->dirty_rx % RX_RING_SIZE].skb == NULL)
refill_rx_ring(dev);
/* If RX ring is not full we are still out of memory.
Restart the timer again. Else we re-add ourselves
to the master poll list.
*/
if (tp->rx_buffers[tp->dirty_rx % RX_RING_SIZE].skb == NULL)
restart_timer();
else netif_rx_schedule(dev); /* we are back on the poll list */
5) dev->close() and dev->suspend() issues
==========================================
The driver writter neednt worry about this. The top net layer takes
care of it.
6) Adding new Stats to /proc
=============================
In order to debug some of the new features, we introduce new stats
that need to be collected.
TODO: Fill this later.
APPENDIX 1: discussion on using ethernet HW FC
==============================================
Most chips with FC only send a pause packet when they run out of Rx buffers.
Since packets are pulled off the DMA ring by a softirq in NAPI,
if the system is slow in grabbing them and we have a high input
rate (faster than the system's capacity to remove packets), then theoretically
there will only be one rx interrupt for all packets during a given packetstorm.
Under low load, we might have a single interrupt per packet.
FC should be programmed to apply in the case when the system cant pull out
packets fast enough i.e send a pause only when you run out of rx buffers.
Note FC in itself is a good solution but we have found it to not be
much of a commodity feature (both in NICs and switches) and hence falls
under the same category as using NIC based mitigation. Also experiments
indicate that its much harder to resolve the resource allocation
issue (aka lazy receiving that NAPI offers) and hence quantify its usefullness
proved harder. In any case, FC works even better with NAPI but is not
necessary.
APPENDIX 2: the "rotting packet" race-window avoidance scheme
=============================================================
There are two types of associations seen here
1) status/int which honors level triggered IRQ
If a status bit for receive or rxnobuff is set and the corresponding
interrupt-enable bit is not on, then no interrupts will be generated. However,
as soon as the "interrupt-enable" bit is unmasked, an immediate interrupt is
generated. [assuming the status bit was not turned off].
Generally the concept of level triggered IRQs in association with a status and
interrupt-enable CSR register set is used to avoid the race.
If we take the example of the tulip:
"pending work" is indicated by the status bit(CSR5 in tulip).
the corresponding interrupt bit (CSR7 in tulip) might be turned off (but
the CSR5 will continue to be turned on with new packet arrivals even if
we clear it the first time)
Very important is the fact that if we turn on the interrupt bit on when
status is set that an immediate irq is triggered.
If we cleared the rx ring and proclaimed there was "no more work
to be done" and then went on to do a few other things; then when we enable
interrupts, there is a possibility that a new packet might sneak in during
this phase. It helps to look at the pseudo code for the tulip poll
routine:
--------------------------
do {
ACK;
while (ring_is_not_empty()) {
work-work-work
if quota is exceeded: exit, no touching irq status/mask
}
/* No packets, but new can arrive while we are doing this*/
CSR5 := read
if (CSR5 is not set) {
/* If something arrives in this narrow window here,
* where the comments are ;-> irq will be generated */
unmask irqs;
exit poll;
}
} while (rx_status_is_set);
------------------------
CSR5 bit of interest is only the rx status.
If you look at the last if statement:
you just finished grabbing all the packets from the rx ring .. you check if
status bit says theres more packets just in ... it says none; you then
enable rx interrupts again; if a new packet just came in during this check,
we are counting that CSR5 will be set in that small window of opportunity
and that by re-enabling interrupts, we would actually triger an interrupt
to register the new packet for processing.
[The above description nay be very verbose, if you have better wording
that will make this more understandable, please suggest it.]
2) non-capable hardware
These do not generally respect level triggered IRQs. Normally,
irqs may be lost while being masked and the only way to leave poll is to do
a double check for new input after netif_rx_complete() is invoked
and re-enable polling (after seeing this new input).
Sample code:
---------
.
.
restart_poll:
while (ring_is_not_empty()) {
work-work-work
if quota is exceeded: exit, not touching irq status/mask
}
.
.
.
enable_rx_interrupts()
netif_rx_complete(dev);
if (ring_has_new_packet() && netif_rx_reschedule(dev, received)) {
disable_rx_and_rxnobufs()
goto restart_poll
} while (rx_status_is_set);
---------
Basically netif_rx_complete() removes us from the poll list, but because a
new packet which will never be caught due to the possibility of a race
might come in, we attempt to re-add ourselves to the poll list.
--------------------------------------------------------------------
relevant sites:
==================
ftp://robur.slu.se/pub/Linux/net-development/NAPI/
--------------------------------------------------------------------
TODO: Write net-skeleton.c driver.
-------------------------------------------------------------
Authors:
========
Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Jamal Hadi Salim <hadi@cyberus.ca>
Robert Olsson <Robert.Olsson@data.slu.se>
Acknowledgements:
================
People who made this document better:
Lennert Buytenhek <buytenh@gnu.org>
Andrew Morton <akpm@zip.com.au>
Manfred Spraul <manfred@colorfullife.com>
Donald Becker <becker@scyld.com>
Jeff Garzik <jgarzik@mandrakesoft.com>
include/linux/netdevice.h
View file @ 4ce937cb
......@@ -206,7 +206,8 @@ enum netdev_state_t
__LINK_STATE_START,
__LINK_STATE_PRESENT,
__LINK_STATE_SCHED,
__LINK_STATE_NOCARRIER
__LINK_STATE_NOCARRIER,
__LINK_STATE_RX_SCHED
};
......@@ -330,6 +331,10 @@ struct net_device
void *ip6_ptr; /* IPv6 specific data */
void *ec_ptr; /* Econet specific data */
struct list_head poll_list; /* Link to poll list */
int quota;
int weight;
struct Qdisc *qdisc;
struct Qdisc *qdisc_sleeping;
struct Qdisc *qdisc_list;
......@@ -373,6 +378,7 @@ struct net_device
int (*stop)(struct net_device *dev);
int (*hard_start_xmit) (struct sk_buff *skb,
struct net_device *dev);
int (*poll) (struct net_device *dev, int *quota);
int (*hard_header) (struct sk_buff *skb,
struct net_device *dev,
unsigned short type,
......@@ -492,8 +498,11 @@ struct softnet_data
int cng_level;
int avg_blog;
struct sk_buff_head input_pkt_queue;
struct list_head poll_list;
struct net_device *output_queue;
struct sk_buff *completion_queue;
struct net_device backlog_dev; /* Sorry. 8) */
} __attribute__((__aligned__(SMP_CACHE_BYTES)));
......@@ -547,6 +556,7 @@ static inline int netif_running(struct net_device *dev)
return test_bit(__LINK_STATE_START, &dev->state);
}
/* Use this variant when it is known for sure that it
* is executing from interrupt context.
*/
......@@ -578,6 +588,8 @@ static inline void dev_kfree_skb_any(struct sk_buff *skb)
extern void net_call_rx_atomic(void (*fn)(void));
#define HAVE_NETIF_RX 1
extern int netif_rx(struct sk_buff *skb);
#define HAVE_NETIF_RECEIVE_SKB 1
extern int netif_receive_skb(struct sk_buff *skb);
extern int dev_ioctl(unsigned int cmd, void *);
extern int dev_change_flags(struct net_device *, unsigned);
extern void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev);
......@@ -695,6 +707,78 @@ enum {
#define netif_msg_rx_status(p) ((p)->msg_enable & NETIF_MSG_RX_STATUS)
#define netif_msg_pktdata(p) ((p)->msg_enable & NETIF_MSG_PKTDATA)
/* Schedule rx intr now? */
static inline int netif_rx_schedule_prep(struct net_device *dev)
{
return netif_running(dev) &&
!test_and_set_bit(__LINK_STATE_RX_SCHED, &dev->state);
}
/* Add interface to tail of rx poll list. This assumes that _prep has
* already been called and returned 1.
*/
static inline void __netif_rx_schedule(struct net_device *dev)
{
unsigned long flags;
int cpu = smp_processor_id();
local_irq_save(flags);
dev_hold(dev);
list_add_tail(&dev->poll_list, &softnet_data[cpu].poll_list);
if (dev->quota < 0)
dev->quota += dev->weight;
else
dev->quota = dev->weight;
__cpu_raise_softirq(cpu, NET_RX_SOFTIRQ);
local_irq_restore(flags);
}
/* Try to reschedule poll. Called by irq handler. */
static inline void netif_rx_schedule(struct net_device *dev)
{
if (netif_rx_schedule_prep(dev))
__netif_rx_schedule(dev);
}
/* Try to reschedule poll. Called by dev->poll() after netif_rx_complete().
* Do not inline this?
*/
static inline int netif_rx_reschedule(struct net_device *dev, int undo)
{
if (netif_rx_schedule_prep(dev)) {
unsigned long flags;
int cpu = smp_processor_id();
dev->quota += undo;
local_irq_save(flags);
list_add_tail(&dev->poll_list, &softnet_data[cpu].poll_list);
__cpu_raise_softirq(cpu, NET_RX_SOFTIRQ);
local_irq_restore(flags);
return 1;
}
return 0;
}
/* Remove interface from poll list: it must be in the poll list
* on current cpu. This primitive is called by dev->poll(), when
* it completes the work. The device cannot be out of poll list at this
* moment, it is BUG().
*/
static inline void netif_rx_complete(struct net_device *dev)
{
unsigned long flags;
local_irq_save(flags);
if (!test_bit(__LINK_STATE_RX_SCHED, &dev->state)) BUG();
list_del(&dev->poll_list);
clear_bit(__LINK_STATE_RX_SCHED, &dev->state);
local_irq_restore(flags);
}
/* These functions live elsewhere (drivers/net/net_init.c, but related) */
extern void ether_setup(struct net_device *dev);
......@@ -719,6 +803,7 @@ extern void dev_mcast_init(void);
extern int netdev_register_fc(struct net_device *dev, void (*stimul)(struct net_device *dev));
extern void netdev_unregister_fc(int bit);
extern int netdev_max_backlog;
extern int weight_p;
extern unsigned long netdev_fc_xoff;
extern atomic_t netdev_dropping;
extern int netdev_set_master(struct net_device *dev, struct net_device *master);
......
include/linux/netfilter_arp.h 0 → 100644
View file @ 4ce937cb
#ifndef __LINUX_ARP_NETFILTER_H
#define __LINUX_ARP_NETFILTER_H
/* ARP-specific defines for netfilter.
* (C)2002 Rusty Russell IBM -- This code is GPL.
*/
#include <linux/config.h>
#include <linux/netfilter.h>
/* There is no PF_ARP. */
#define NF_ARP 0
/* ARP Hooks */
#define NF_ARP_IN 0
#define NF_ARP_OUT 1
#define NF_ARP_NUMHOOKS 2
#endif /* __LINUX_ARP_NETFILTER_H */
include/linux/netfilter_arp/arp_tables.h 0 → 100644
View file @ 4ce937cb
/*
* Format of an ARP firewall descriptor
*
* src, tgt, src_mask, tgt_mask, arpop, arpop_mask are always stored in
* network byte order.
* flags are stored in host byte order (of course).
*/
#ifndef _ARPTABLES_H
#define _ARPTABLES_H
#ifdef __KERNEL__
#include <linux/if.h>
#include <linux/types.h>
#include <linux/in.h>
#include <linux/if_arp.h>
#include <linux/skbuff.h>
#endif
#include <linux/netfilter_arp.h>
#define ARPT_FUNCTION_MAXNAMELEN 30
#define ARPT_TABLE_MAXNAMELEN 32
#define ARPT_DEV_ADDR_LEN_MAX 16
struct arpt_devaddr_info {
char addr[ARPT_DEV_ADDR_LEN_MAX];
char mask[ARPT_DEV_ADDR_LEN_MAX];
};
/* Yes, Virginia, you have to zero the padding. */
struct arpt_arp {
/* Source and target IP addr */
struct in_addr src, tgt;
/* Mask for src and target IP addr */
struct in_addr smsk, tmsk;
/* Device hw address length, src+target device addresses */
u_int8_t arhln, arhln_mask;
struct arpt_devaddr_info src_devaddr;
struct arpt_devaddr_info tgt_devaddr;
/* ARP operation code. */
u_int16_t arpop, arpop_mask;
/* ARP hardware address and protocol address format. */
u_int16_t arhrd, arhrd_mask;
u_int16_t arpro, arpro_mask;
/* The protocol address length is only accepted if it is 4
* so there is no use in offering a way to do filtering on it.
*/
char iniface[IFNAMSIZ], outiface[IFNAMSIZ];
unsigned char iniface_mask[IFNAMSIZ], outiface_mask[IFNAMSIZ];
/* Flags word */
u_int8_t flags;
/* Inverse flags */
u_int16_t invflags;
};
struct arpt_entry_target
{
union {
struct {
u_int16_t target_size;
/* Used by userspace */
char name[ARPT_FUNCTION_MAXNAMELEN];
} user;
struct {
u_int16_t target_size;
/* Used inside the kernel */
struct arpt_target *target;
} kernel;
/* Total length */
u_int16_t target_size;
} u;
unsigned char data[0];
};
struct arpt_standard_target
{
struct arpt_entry_target target;
int verdict;
};
struct arpt_counters
{
u_int64_t pcnt, bcnt; /* Packet and byte counters */
};
/* Values for "flag" field in struct arpt_ip (general arp structure).
* No flags defined yet.
*/
#define ARPT_F_MASK 0x00 /* All possible flag bits mask. */
/* Values for "inv" field in struct arpt_arp. */
#define ARPT_INV_VIA_IN 0x0001 /* Invert the sense of IN IFACE. */
#define ARPT_INV_VIA_OUT 0x0002 /* Invert the sense of OUT IFACE */
#define ARPT_INV_SRCIP 0x0004 /* Invert the sense of SRC IP. */
#define ARPT_INV_TGTIP 0x0008 /* Invert the sense of TGT IP. */
#define ARPT_INV_SRCDEVADDR 0x0010 /* Invert the sense of SRC DEV ADDR. */
#define ARPT_INV_TGTDEVADDR 0x0020 /* Invert the sense of TGT DEV ADDR. */
#define ARPT_INV_ARPOP 0x0040 /* Invert the sense of ARP OP. */
#define ARPT_INV_ARPHRD 0x0080 /* Invert the sense of ARP HRD. */
#define ARPT_INV_ARPPRO 0x0100 /* Invert the sense of ARP PRO. */
#define ARPT_INV_ARPHLN 0x0200 /* Invert the sense of ARP HLN. */
#define ARPT_INV_MASK 0x007F /* All possible flag bits mask. */
/* This structure defines each of the firewall rules. Consists of 3
parts which are 1) general ARP header stuff 2) match specific
stuff 3) the target to perform if the rule matches */
struct arpt_entry
{
struct arpt_arp arp;
/* Size of arpt_entry + matches */
u_int16_t target_offset;
/* Size of arpt_entry + matches + target */
u_int16_t next_offset;
/* Back pointer */
unsigned int comefrom;
/* Packet and byte counters. */
struct arpt_counters counters;
/* The matches (if any), then the target. */
unsigned char elems[0];
};
/*
* New IP firewall options for [gs]etsockopt at the RAW IP level.
* Unlike BSD Linux inherits IP options so you don't have to use a raw
* socket for this. Instead we check rights in the calls.
*/
#define ARPT_BASE_CTL 96 /* base for firewall socket options */
#define ARPT_SO_SET_REPLACE (ARPT_BASE_CTL)
#define ARPT_SO_SET_ADD_COUNTERS (ARPT_BASE_CTL + 1)
#define ARPT_SO_SET_MAX ARPT_SO_SET_ADD_COUNTERS
#define ARPT_SO_GET_INFO (ARPT_BASE_CTL)
#define ARPT_SO_GET_ENTRIES (ARPT_BASE_CTL + 1)
#define ARPT_SO_GET_MAX ARPT_SO_GET_ENTRIES
/* CONTINUE verdict for targets */
#define ARPT_CONTINUE 0xFFFFFFFF
/* For standard target */
#define ARPT_RETURN (-NF_MAX_VERDICT - 1)
/* The argument to ARPT_SO_GET_INFO */
struct arpt_getinfo
{
/* Which table: caller fills this in. */
char name[ARPT_TABLE_MAXNAMELEN];
/* Kernel fills these in. */
/* Which hook entry points are valid: bitmask */
unsigned int valid_hooks;
/* Hook entry points: one per netfilter hook. */
unsigned int hook_entry[NF_ARP_NUMHOOKS];
/* Underflow points. */
unsigned int underflow[NF_ARP_NUMHOOKS];
/* Number of entries */
unsigned int num_entries;
/* Size of entries. */
unsigned int size;
};
/* The argument to ARPT_SO_SET_REPLACE. */
struct arpt_replace
{
/* Which table. */
char name[ARPT_TABLE_MAXNAMELEN];
/* Which hook entry points are valid: bitmask. You can't
change this. */
unsigned int valid_hooks;
/* Number of entries */
unsigned int num_entries;
/* Total size of new entries */
unsigned int size;
/* Hook entry points. */
unsigned int hook_entry[NF_ARP_NUMHOOKS];
/* Underflow points. */
unsigned int underflow[NF_ARP_NUMHOOKS];
/* Information about old entries: */
/* Number of counters (must be equal to current number of entries). */
unsigned int num_counters;
/* The old entries' counters. */
struct arpt_counters *counters;
/* The entries (hang off end: not really an array). */
struct arpt_entry entries[0];
};
/* The argument to ARPT_SO_ADD_COUNTERS. */
struct arpt_counters_info
{
/* Which table. */
char name[ARPT_TABLE_MAXNAMELEN];
unsigned int num_counters;
/* The counters (actually `number' of these). */
struct arpt_counters counters[0];
};
/* The argument to ARPT_SO_GET_ENTRIES. */
struct arpt_get_entries
{
/* Which table: user fills this in. */
char name[ARPT_TABLE_MAXNAMELEN];
/* User fills this in: total entry size. */
unsigned int size;
/* The entries. */
struct arpt_entry entrytable[0];
};
/* Standard return verdict, or do jump. */
#define ARPT_STANDARD_TARGET ""
/* Error verdict. */
#define ARPT_ERROR_TARGET "ERROR"
/* Helper functions */
static __inline__ struct arpt_entry_target *arpt_get_target(struct arpt_entry *e)
{
return (void *)e + e->target_offset;
}
/* fn returns 0 to continue iteration */
#define ARPT_ENTRY_ITERATE(entries, size, fn, args...) \
({ \
unsigned int __i; \
int __ret = 0; \
struct arpt_entry *__entry; \
\
for (__i = 0; __i < (size); __i += __entry->next_offset) { \
__entry = (void *)(entries) + __i; \
\
__ret = fn(__entry , ## args); \
if (__ret != 0) \
break; \
} \
__ret; \
})
/*
* Main firewall chains definitions and global var's definitions.
*/
#ifdef __KERNEL__
/* Registration hooks for targets. */
struct arpt_target
{
struct list_head list;
const char name[ARPT_FUNCTION_MAXNAMELEN];
/* Returns verdict. */
unsigned int (*target)(struct sk_buff **pskb,
unsigned int hooknum,
const struct net_device *in,
const struct net_device *out,
const void *targinfo,
void *userdata);
/* Called when user tries to insert an entry of this type:
hook_mask is a bitmask of hooks from which it can be
called. */
/* Should return true or false. */
int (*checkentry)(const char *tablename,
const struct arpt_entry *e,
void *targinfo,
unsigned int targinfosize,
unsigned int hook_mask);
/* Called when entry of this type deleted. */
void (*destroy)(void *targinfo, unsigned int targinfosize);
/* Set this to THIS_MODULE if you are a module, otherwise NULL */
struct module *me;
};
extern int arpt_register_target(struct arpt_target *target);
extern void arpt_unregister_target(struct arpt_target *target);
/* Furniture shopping... */
struct arpt_table
{
struct list_head list;
/* A unique name... */
char name[ARPT_TABLE_MAXNAMELEN];
/* Seed table: copied in register_table */
struct arpt_replace *table;
/* What hooks you will enter on */
unsigned int valid_hooks;
/* Lock for the curtain */
rwlock_t lock;
/* Man behind the curtain... */
struct arpt_table_info *private;
/* Set this to THIS_MODULE if you are a module, otherwise NULL */
struct module *me;
};
extern int arpt_register_table(struct arpt_table *table);
extern void arpt_unregister_table(struct arpt_table *table);
extern unsigned int arpt_do_table(struct sk_buff **pskb,
unsigned int hook,
const struct net_device *in,
const struct net_device *out,
struct arpt_table *table,
void *userdata);
#define ARPT_ALIGN(s) (((s) + (__alignof__(struct arpt_entry)-1)) & ~(__alignof__(struct arpt_entry)-1))
#endif /*__KERNEL__*/
#endif /* _ARPTABLES_H */
include/linux/netfilter_ipv4/ip_nat.h
View file @ 4ce937cb
......@@ -11,8 +11,13 @@ enum ip_nat_manip_type
IP_NAT_MANIP_DST
};
#ifndef CONFIG_IP_NF_NAT_LOCAL
/* SRC manip occurs only on POST_ROUTING */
#define HOOK2MANIP(hooknum) ((hooknum) != NF_IP_POST_ROUTING)
#else
/* SRC manip occurs POST_ROUTING or LOCAL_IN */
#define HOOK2MANIP(hooknum) ((hooknum) != NF_IP_POST_ROUTING && (hooknum) != NF_IP_LOCAL_IN)
#endif
/* 2.3.19 (I hope) will define this in linux/netfilter_ipv4.h. */
#ifndef SO_ORIGINAL_DST
......
include/linux/sysctl.h
View file @ 4ce937cb
......@@ -202,7 +202,8 @@ enum
NET_CORE_NO_CONG_THRESH=13,
NET_CORE_NO_CONG=14,
NET_CORE_LO_CONG=15,
NET_CORE_MOD_CONG=16
NET_CORE_MOD_CONG=16,
NET_CORE_DEV_WEIGHT=17
};
/* /proc/sys/net/ethernet */
......
net/core/dev.c
View file @ 4ce937cb
......@@ -798,6 +798,19 @@ int dev_close(struct net_device *dev)
clear_bit(__LINK_STATE_START, &dev->state);
/* Synchronize to scheduled poll. We cannot touch poll list,
* it can be even on different cpu. So just clear netif_running(),
* and wait when poll really will happen. Actually, the best place
* for this is inside dev->stop() after device stopped its irq
* engine, but this requires more changes in devices. */
smp_mb__after_clear_bit(); /* Commit netif_running(). */
while (test_bit(__LINK_STATE_RX_SCHED, &dev->state)) {
/* No hurry. */
current->state = TASK_INTERRUPTIBLE;
schedule_timeout(1);
}
/*
* Call the device specific close. This cannot fail.
* Only if device is UP
......@@ -1072,6 +1085,7 @@ int dev_queue_xmit(struct sk_buff *skb)
=======================================================================*/
int netdev_max_backlog = 300;
int weight_p = 64; /* old backlog weight */
/* These numbers are selected based on intuition and some
* experimentatiom, if you have more scientific way of doing this
* please go ahead and fix things.
......@@ -1237,13 +1251,11 @@ int netif_rx(struct sk_buff *skb)
enqueue:
dev_hold(skb->dev);
__skb_queue_tail(&queue->input_pkt_queue,skb);
/* Runs from irqs or BH's, no need to wake BH */
cpu_raise_softirq(this_cpu, NET_RX_SOFTIRQ);
local_irq_restore(flags);
#ifndef OFFLINE_SAMPLE
get_sample_stats(this_cpu);
#endif
return softnet_data[this_cpu].cng_level;
return queue->cng_level;
}
if (queue->throttle) {
......@@ -1253,6 +1265,8 @@ int netif_rx(struct sk_buff *skb)
netdev_wakeup();
#endif
}
netif_rx_schedule(&queue->backlog_dev);
goto enqueue;
}
......@@ -1308,19 +1322,12 @@ static int deliver_to_old_ones(struct packet_type *pt, struct sk_buff *skb, int
return ret;
}
/* Reparent skb to master device. This function is called
* only from net_rx_action under BR_NETPROTO_LOCK. It is misuse
* of BR_NETPROTO_LOCK, but it is OK for now.
*/
static __inline__ void skb_bond(struct sk_buff *skb)
{
struct net_device *dev = skb->dev;
if (dev->master) {
dev_hold(dev->master);
if (dev->master)
skb->dev = dev->master;
dev_put(dev);
}
}
static void net_tx_action(struct softirq_action *h)
......@@ -1416,121 +1423,138 @@ static inline void handle_diverter(struct sk_buff *skb)
}
#endif /* CONFIG_NET_DIVERT */
static void net_rx_action(struct softirq_action *h)
int netif_receive_skb(struct sk_buff *skb)
{
int this_cpu = smp_processor_id();
struct softnet_data *queue = &softnet_data[this_cpu];
unsigned long start_time = jiffies;
int bugdet = netdev_max_backlog;
br_read_lock(BR_NETPROTO_LOCK);
for (;;) {
struct sk_buff *skb;
struct net_device *rx_dev;
local_irq_disable();
skb = __skb_dequeue(&queue->input_pkt_queue);
local_irq_enable();
struct packet_type *ptype, *pt_prev;
int ret = NET_RX_DROP;
unsigned short type = skb->protocol;
if (skb == NULL)
break;
if (skb->stamp.tv_sec == 0)
do_gettimeofday(&skb->stamp);
skb_bond(skb);
skb_bond(skb);
rx_dev = skb->dev;
netdev_rx_stat[smp_processor_id()].total++;
#ifdef CONFIG_NET_FASTROUTE
if (skb->pkt_type == PACKET_FASTROUTE) {
netdev_rx_stat[this_cpu].fastroute_deferred_out++;
dev_queue_xmit(skb);
dev_put(rx_dev);
continue;
}
if (skb->pkt_type == PACKET_FASTROUTE) {
netdev_rx_stat[smp_processor_id()].fastroute_deferred_out++;
return dev_queue_xmit(skb);
}
#endif
skb->h.raw = skb->nh.raw = skb->data;
{
struct packet_type *ptype, *pt_prev;
unsigned short type = skb->protocol;
pt_prev = NULL;
for (ptype = ptype_all; ptype; ptype = ptype->next) {
if (!ptype->dev || ptype->dev == skb->dev) {
if (pt_prev) {
if (!pt_prev->data) {
deliver_to_old_ones(pt_prev, skb, 0);
} else {
atomic_inc(&skb->users);
pt_prev->func(skb,
skb->dev,
pt_prev);
}
}
pt_prev = ptype;
skb->h.raw = skb->nh.raw = skb->data;
pt_prev = NULL;
for (ptype = ptype_all; ptype; ptype = ptype->next) {
if (!ptype->dev || ptype->dev == skb->dev) {
if (pt_prev) {
if (!pt_prev->data) {
ret = deliver_to_old_ones(pt_prev, skb, 0);
} else {
atomic_inc(&skb->users);
ret = pt_prev->func(skb, skb->dev, pt_prev);
}
}
pt_prev = ptype;
}
}
#ifdef CONFIG_NET_DIVERT
if (skb->dev->divert && skb->dev->divert->divert)
handle_diverter(skb);
if (skb->dev->divert && skb->dev->divert->divert)
ret = handle_diverter(skb);
#endif /* CONFIG_NET_DIVERT */
#if defined(CONFIG_BRIDGE) || defined(CONFIG_BRIDGE_MODULE)
if (skb->dev->br_port != NULL &&
br_handle_frame_hook != NULL) {
handle_bridge(skb, pt_prev);
dev_put(rx_dev);
continue;
}
if (skb->dev->br_port != NULL &&
br_handle_frame_hook != NULL) {
return handle_bridge(skb, pt_prev);
}
#endif
for (ptype=ptype_base[ntohs(type)&15];ptype;ptype=ptype->next) {
if (ptype->type == type &&
(!ptype->dev || ptype->dev == skb->dev)) {
if (pt_prev) {
if (!pt_prev->data)
deliver_to_old_ones(pt_prev, skb, 0);
else {
atomic_inc(&skb->users);
pt_prev->func(skb,
skb->dev,
pt_prev);
}
}
pt_prev = ptype;
for (ptype=ptype_base[ntohs(type)&15];ptype;ptype=ptype->next) {
if (ptype->type == type &&
(!ptype->dev || ptype->dev == skb->dev)) {
if (pt_prev) {
if (!pt_prev->data) {
ret = deliver_to_old_ones(pt_prev, skb, 0);
} else {
atomic_inc(&skb->users);
ret = pt_prev->func(skb, skb->dev, pt_prev);
}
}
pt_prev = ptype;
}
}
if (pt_prev) {
if (!pt_prev->data)
deliver_to_old_ones(pt_prev, skb, 1);
else
pt_prev->func(skb, skb->dev, pt_prev);
} else
kfree_skb(skb);
if (pt_prev) {
if (!pt_prev->data) {
ret = deliver_to_old_ones(pt_prev, skb, 1);
} else {
ret = pt_prev->func(skb, skb->dev, pt_prev);
}
} else {
kfree_skb(skb);
/* Jamal, now you will not able to escape explaining
* me how you were going to use this. :-)
*/
ret = NET_RX_DROP;
}
dev_put(rx_dev);
return ret;
}
if (bugdet-- < 0 || jiffies - start_time > 1)
goto softnet_break;
static int process_backlog(struct net_device *backlog_dev, int *budget)
{
int work = 0;
int quota = min(backlog_dev->quota, *budget);
int this_cpu = smp_processor_id();
struct softnet_data *queue = &softnet_data[this_cpu];
unsigned long start_time = jiffies;
for (;;) {
struct sk_buff *skb;
struct net_device *dev;
local_irq_disable();
skb = __skb_dequeue(&queue->input_pkt_queue);
if (skb == NULL)
goto job_done;
local_irq_enable();
dev = skb->dev;
netif_receive_skb(skb);
dev_put(dev);
work++;
if (work >= quota || jiffies - start_time > 1)
break;
#ifdef CONFIG_NET_HW_FLOWCONTROL
if (queue->throttle && queue->input_pkt_queue.qlen < no_cong_thresh ) {
if (atomic_dec_and_test(&netdev_dropping)) {
queue->throttle = 0;
netdev_wakeup();
goto softnet_break;
if (queue->throttle && queue->input_pkt_queue.qlen < no_cong_thresh ) {
if (atomic_dec_and_test(&netdev_dropping)) {
queue->throttle = 0;
netdev_wakeup();
break;
}
}
}
#endif
}
br_read_unlock(BR_NETPROTO_LOCK);
local_irq_disable();
backlog_dev->quota -= work;
*budget -= work;
return -1;
job_done:
backlog_dev->quota -= work;
*budget -= work;
list_del(&backlog_dev->poll_list);
clear_bit(__LINK_STATE_RX_SCHED, &backlog_dev->state);
if (queue->throttle) {
queue->throttle = 0;
#ifdef CONFIG_NET_HW_FLOWCONTROL
......@@ -1539,21 +1563,53 @@ static void net_rx_action(struct softirq_action *h)
#endif
}
local_irq_enable();
return 0;
}
NET_PROFILE_LEAVE(softnet_process);
return;
static void net_rx_action(struct softirq_action *h)
{
int this_cpu = smp_processor_id();
struct softnet_data *queue = &softnet_data[this_cpu];
unsigned long start_time = jiffies;
int budget = netdev_max_backlog;
softnet_break:
br_read_lock(BR_NETPROTO_LOCK);
local_irq_disable();
while (!list_empty(&queue->poll_list)) {
struct net_device *dev;
if (budget <= 0 || jiffies - start_time > 1)
goto softnet_break;
local_irq_enable();
dev = list_entry(queue->poll_list.next, struct net_device, poll_list);
if (dev->quota <= 0 || dev->poll(dev, &budget)) {
local_irq_disable();
list_del(&dev->poll_list);
list_add_tail(&dev->poll_list, &queue->poll_list);
if (dev->quota < 0)
dev->quota += dev->weight;
else
dev->quota = dev->weight;
} else {
dev_put(dev);
local_irq_disable();
}
}
local_irq_enable();
br_read_unlock(BR_NETPROTO_LOCK);
return;
local_irq_disable();
softnet_break:
netdev_rx_stat[this_cpu].time_squeeze++;
/* This already runs in BH context, no need to wake up BH's */
cpu_raise_softirq(this_cpu, NET_RX_SOFTIRQ);
local_irq_enable();
__cpu_raise_softirq(this_cpu, NET_RX_SOFTIRQ);
NET_PROFILE_LEAVE(softnet_process);
return;
local_irq_enable();
br_read_unlock(BR_NETPROTO_LOCK);
}
static gifconf_func_t * gifconf_list [NPROTO];
......@@ -2626,6 +2682,7 @@ int __init net_dev_init(void)
if (!dev_boot_phase)
return 0;
#ifdef CONFIG_NET_DIVERT
dv_init();
#endif /* CONFIG_NET_DIVERT */
......@@ -2643,8 +2700,13 @@ int __init net_dev_init(void)
queue->cng_level = 0;
queue->avg_blog = 10; /* arbitrary non-zero */
queue->completion_queue = NULL;
INIT_LIST_HEAD(&queue->poll_list);
set_bit(__LINK_STATE_START, &queue->backlog_dev.state);
queue->backlog_dev.weight = weight_p;
queue->backlog_dev.poll = process_backlog;
atomic_set(&queue->backlog_dev.refcnt, 1);
}
#ifdef CONFIG_NET_PROFILE
net_profile_init();
NET_PROFILE_REGISTER(dev_queue_xmit);
......@@ -2744,7 +2806,6 @@ int __init net_dev_init(void)
#ifdef CONFIG_NET_SCHED
pktsched_init();
#endif
/*
* Initialise network devices
*/
......
net/core/sysctl_net_core.c
View file @ 4ce937cb
......@@ -12,6 +12,7 @@
#ifdef CONFIG_SYSCTL
extern int netdev_max_backlog;
extern int weight_p;
extern int no_cong_thresh;
extern int no_cong;
extern int lo_cong;
......@@ -47,6 +48,9 @@ ctl_table core_table[] = {
{NET_CORE_RMEM_DEFAULT, "rmem_default",
&sysctl_rmem_default, sizeof(int), 0644, NULL,
&proc_dointvec},
{NET_CORE_DEV_WEIGHT, "dev_weight",
&weight_p, sizeof(int), 0644, NULL,
&proc_dointvec},
{NET_CORE_MAX_BACKLOG, "netdev_max_backlog",
&netdev_max_backlog, sizeof(int), 0644, NULL,
&proc_dointvec},
......
net/ipv4/arp.c
View file @ 4ce937cb
......@@ -112,7 +112,7 @@
#include <asm/system.h>
#include <asm/uaccess.h>
#include <linux/netfilter_arp.h>
/*
* Interface to generic neighbour cache.
......@@ -561,7 +561,8 @@ void arp_send(int type, int ptype, u32 dest_ip,
arp_ptr+=dev->addr_len;
memcpy(arp_ptr, &dest_ip, 4);
dev_queue_xmit(skb);
/* Send it off, maybe filter it using firewalling first. */
NF_HOOK(NF_ARP, NF_ARP_OUT, skb, NULL, dev, dev_queue_xmit);
return;
out:
......@@ -574,45 +575,31 @@ static void parp_redo(struct sk_buff *skb)
}
/*
* Receive an arp request by the device layer.
* Process an arp request.
*/
int arp_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt)
int arp_process(struct sk_buff *skb)
{
struct arphdr *arp = skb->nh.arph;
unsigned char *arp_ptr= (unsigned char *)(arp+1);
struct net_device *dev = skb->dev;
struct in_device *in_dev = in_dev_get(dev);
struct arphdr *arp;
unsigned char *arp_ptr;
struct rtable *rt;
unsigned char *sha, *tha;
u32 sip, tip;
u16 dev_type = dev->type;
int addr_type;
struct in_device *in_dev = in_dev_get(dev);
struct neighbour *n;
/*
* The hardware length of the packet should match the hardware length
* of the device. Similarly, the hardware types should match. The
* device should be ARP-able. Also, if pln is not 4, then the lookup
* is not from an IP number. We can't currently handle this, so toss
* it.
*/
if (in_dev == NULL ||
arp->ar_hln != dev->addr_len ||
dev->flags & IFF_NOARP ||
skb->pkt_type == PACKET_OTHERHOST ||
skb->pkt_type == PACKET_LOOPBACK ||
arp->ar_pln != 4)
/* arp_rcv below verifies the ARP header, verifies the device
* is ARP'able, and linearizes the SKB (if needed).
*/
if (in_dev == NULL)
goto out;
if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL)
goto out_of_mem;
if (skb_is_nonlinear(skb)) {
if (skb_linearize(skb, GFP_ATOMIC) != 0)
goto freeskb;
arp = skb->nh.arph;
arp_ptr= (unsigned char *)(arp+1);
}
arp = skb->nh.arph;
arp_ptr= (unsigned char *)(arp+1);
switch (dev_type) {
default:
......@@ -827,13 +814,41 @@ int arp_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt)
out:
if (in_dev)
in_dev_put(in_dev);
freeskb:
kfree_skb(skb);
out_of_mem:
return 0;
}
/*
* Receive an arp request from the device layer.
*/
int arp_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt)
{
struct arphdr *arp = skb->nh.arph;
if (arp->ar_hln != dev->addr_len ||
dev->flags & IFF_NOARP ||
skb->pkt_type == PACKET_OTHERHOST ||
skb->pkt_type == PACKET_LOOPBACK ||
arp->ar_pln != 4)
goto freeskb;
if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL)
goto out_of_mem;
if (skb_is_nonlinear(skb)) {
if (skb_linearize(skb, GFP_ATOMIC) != 0)
goto freeskb;
}
return NF_HOOK(NF_ARP, NF_ARP_IN, skb, dev, NULL, arp_process);
freeskb:
kfree_skb(skb);
out_of_mem:
return 0;
}
/*
* User level interface (ioctl, /proc)
......
net/ipv4/netfilter/Config.in
View file @ 4ce937cb
......@@ -47,6 +47,7 @@ if [ "$CONFIG_IP_NF_IPTABLES" != "n" ]; then
define_bool CONFIG_IP_NF_NAT_NEEDED y
dep_tristate ' MASQUERADE target support' CONFIG_IP_NF_TARGET_MASQUERADE $CONFIG_IP_NF_NAT
dep_tristate ' REDIRECT target support' CONFIG_IP_NF_TARGET_REDIRECT $CONFIG_IP_NF_NAT
bool ' NAT of local connections' CONFIG_IP_NF_NAT_LOCAL
if [ "$CONFIG_EXPERIMENTAL" = "y" ]; then
dep_tristate ' Basic SNMP-ALG support (EXPERIMENTAL)' CONFIG_IP_NF_NAT_SNMP_BASIC $CONFIG_IP_NF_NAT
fi
......@@ -79,6 +80,11 @@ if [ "$CONFIG_IP_NF_IPTABLES" != "n" ]; then
dep_tristate ' TCPMSS target support' CONFIG_IP_NF_TARGET_TCPMSS $CONFIG_IP_NF_IPTABLES
fi
tristate 'ARP tables support' CONFIG_IP_NF_ARPTABLES
if [ "$CONFIG_IP_NF_ARPTABLES" != "n" ]; then
dep_tristate ' ARP packet filtering' CONFIG_IP_NF_ARPFILTER $CONFIG_IP_NF_ARPTABLES
fi
# Backwards compatibility modules: only if you don't build in the others.
if [ "$CONFIG_IP_NF_CONNTRACK" != "y" ]; then
if [ "$CONFIG_IP_NF_IPTABLES" != "y" ]; then
......
net/ipv4/netfilter/Makefile
View file @ 4ce937cb
......@@ -9,7 +9,7 @@
O_TARGET := netfilter.o
export-objs = ip_conntrack_standalone.o ip_conntrack_ftp.o ip_fw_compat.o ip_nat_standalone.o ip_tables.o
export-objs = ip_conntrack_standalone.o ip_conntrack_ftp.o ip_fw_compat.o ip_nat_standalone.o ip_tables.o arp_tables.o
# Multipart objects.
list-multi := ip_conntrack.o iptable_nat.o ipfwadm.o ipchains.o
......@@ -75,6 +75,12 @@ obj-$(CONFIG_IP_NF_TARGET_LOG) += ipt_LOG.o
obj-$(CONFIG_IP_NF_TARGET_ULOG) += ipt_ULOG.o
obj-$(CONFIG_IP_NF_TARGET_TCPMSS) += ipt_TCPMSS.o
# generic ARP tables
obj-$(CONFIG_IP_NF_ARPTABLES) += arp_tables.o
# just filtering instance of ARP tables for now
obj-$(CONFIG_IP_NF_ARPFILTER) += arptable_filter.o
# backwards compatibility
obj-$(CONFIG_IP_NF_COMPAT_IPCHAINS) += ipchains.o
obj-$(CONFIG_IP_NF_COMPAT_IPFWADM) += ipfwadm.o
......
net/ipv4/netfilter/arp_tables.c 0 → 100644
View file @ 4ce937cb
/*
* Packet matching code for ARP packets.
*
* Based heavily, if not almost entirely, upon ip_tables.c framework.
*
* Some ARP specific bits are:
*
* Copyright (C) 2002 David S. Miller (davem@redhat.com)
*
*/
#include <linux/config.h>
#include <linux/kernel.h>
#include <linux/skbuff.h>
#include <linux/netdevice.h>
#include <linux/if_arp.h>
#include <linux/kmod.h>
#include <linux/vmalloc.h>
#include <linux/proc_fs.h>
#include <linux/module.h>
#include <linux/init.h>
#include <asm/uaccess.h>
#include <asm/semaphore.h>
#include <linux/netfilter_arp/arp_tables.h>
/*#define DEBUG_ARP_TABLES*/
/*#define DEBUG_ARP_TABLES_USER*/
#ifdef DEBUG_ARP_TABLES
#define dprintf(format, args...) printk(format , ## args)
#else
#define dprintf(format, args...)
#endif
#ifdef DEBUG_ARP_TABLES_USER
#define duprintf(format, args...) printk(format , ## args)
#else
#define duprintf(format, args...)
#endif
#ifdef CONFIG_NETFILTER_DEBUG
#define ARP_NF_ASSERT(x) \
do { \
if (!(x)) \
printk("ARP_NF_ASSERT: %s:%s:%u\n", \
__FUNCTION__, __FILE__, __LINE__); \
} while(0)
#else
#define ARP_NF_ASSERT(x)
#endif
#define SMP_ALIGN(x) (((x) + SMP_CACHE_BYTES-1) & ~(SMP_CACHE_BYTES-1))
static DECLARE_MUTEX(arpt_mutex);
#define ASSERT_READ_LOCK(x) ARP_NF_ASSERT(down_trylock(&arpt_mutex) != 0)
#define ASSERT_WRITE_LOCK(x) ARP_NF_ASSERT(down_trylock(&arpt_mutex) != 0)
#include <linux/netfilter_ipv4/lockhelp.h>
#include <linux/netfilter_ipv4/listhelp.h>
struct arpt_table_info {
unsigned int size;
unsigned int number;
unsigned int initial_entries;
unsigned int hook_entry[NF_ARP_NUMHOOKS];
unsigned int underflow[NF_ARP_NUMHOOKS];
char entries[0] __attribute__((aligned(SMP_CACHE_BYTES)));
};
static LIST_HEAD(arpt_target);
static LIST_HEAD(arpt_tables);
#define ADD_COUNTER(c,b,p) do { (c).bcnt += (b); (c).pcnt += (p); } while(0)
#ifdef CONFIG_SMP
#define TABLE_OFFSET(t,p) (SMP_ALIGN((t)->size)*(p))
#else
#define TABLE_OFFSET(t,p) 0
#endif
static inline int arp_devaddr_compare(const struct arpt_devaddr_info *ap,
char *hdr_addr, int len)
{
int i, ret;
if (len > ARPT_DEV_ADDR_LEN_MAX)
len = ARPT_DEV_ADDR_LEN_MAX;
ret = 0;
for (i = 0; i < len; i++)
ret |= (hdr_addr[i] ^ ap->addr[i]) & ap->mask[i];
return (ret != 0);
}
/* Returns whether packet matches rule or not. */
static inline int arp_packet_match(const struct arphdr *arphdr,
struct net_device *dev,
const char *indev,
const char *outdev,
const struct arpt_arp *arpinfo)
{
char *arpptr = (char *)(arphdr + 1);
char *src_devaddr, *tgt_devaddr;
u32 *src_ipaddr, *tgt_ipaddr;
int i, ret;
#define FWINV(bool,invflg) ((bool) ^ !!(arpinfo->invflags & invflg))
if (FWINV((arphdr->ar_op & arpinfo->arpop_mask) != arpinfo->arpop,
ARPT_INV_ARPOP)) {
dprintf("ARP operation field mismatch.\n");
dprintf("ar_op: %04x info->arpop: %04x info->arpop_mask: %04x\n",
arphdr->ar_op, arpinfo->arpop, arpinfo->arpop_mask);
return 0;
}
if (FWINV((arphdr->ar_hrd & arpinfo->arhrd_mask) != arpinfo->arhrd,
ARPT_INV_ARPHRD)) {
dprintf("ARP hardware address format mismatch.\n");
dprintf("ar_hrd: %04x info->arhrd: %04x info->arhrd_mask: %04x\n",
arphdr->ar_hrd, arpinfo->arhrd, arpinfo->arhrd_mask);
return 0;
}
if (FWINV((arphdr->ar_pro & arpinfo->arpro_mask) != arpinfo->arpro,
ARPT_INV_ARPPRO)) {
dprintf("ARP protocol address format mismatch.\n");
dprintf("ar_pro: %04x info->arpro: %04x info->arpro_mask: %04x\n",
arphdr->ar_pro, arpinfo->arpro, arpinfo->arpro_mask);
return 0;
}
if (FWINV((arphdr->ar_hln & arpinfo->arhln_mask) != arpinfo->arhln,
ARPT_INV_ARPHLN)) {
dprintf("ARP hardware address length mismatch.\n");
dprintf("ar_hln: %02x info->arhln: %02x info->arhln_mask: %02x\n",
arphdr->ar_hln, arpinfo->arhln, arpinfo->arhln_mask);
}
src_devaddr = arpptr;
arpptr += dev->addr_len;
src_ipaddr = (u32 *) arpptr;
arpptr += sizeof(u32);
tgt_devaddr = arpptr;
arpptr += dev->addr_len;
tgt_ipaddr = (u32 *) arpptr;
if (FWINV(arp_devaddr_compare(&arpinfo->src_devaddr, src_devaddr, dev->addr_len),
ARPT_INV_SRCDEVADDR) ||
FWINV(arp_devaddr_compare(&arpinfo->tgt_devaddr, tgt_devaddr, dev->addr_len),
ARPT_INV_TGTDEVADDR)) {
dprintf("Source or target device address mismatch.\n");
return 0;
}
if (FWINV(((*src_ipaddr) & arpinfo->smsk.s_addr) != arpinfo->src.s_addr,
ARPT_INV_SRCIP) ||
FWINV((((*tgt_ipaddr) & arpinfo->tmsk.s_addr) != arpinfo->tgt.s_addr),
ARPT_INV_TGTIP)) {
dprintf("Source or target IP address mismatch.\n");
dprintf("SRC: %u.%u.%u.%u. Mask: %u.%u.%u.%u. Target: %u.%u.%u.%u.%s\n",
NIPQUAD(*src_ipaddr),
NIPQUAD(arpinfo->smsk.s_addr),
NIPQUAD(arpinfo->src.s_addr),
arpinfo->invflags & ARPT_INV_SRCIP ? " (INV)" : "");
dprintf("TGT: %u.%u.%u.%u Mask: %u.%u.%u.%u Target: %u.%u.%u.%u.%s\n",
NIPQUAD(*tgt_ipaddr),
NIPQUAD(arpinfo->tmsk.s_addr),
NIPQUAD(arpinfo->tgt.s_addr),
arpinfo->invflags & ARPT_INV_TGTIP ? " (INV)" : "");
return 0;
}
/* Look for ifname matches; this should unroll nicely. */
for (i = 0, ret = 0; i < IFNAMSIZ/sizeof(unsigned long); i++) {
ret |= (((const unsigned long *)indev)[i]
^ ((const unsigned long *)arpinfo->iniface)[i])
& ((const unsigned long *)arpinfo->iniface_mask)[i];
}
if (FWINV(ret != 0, ARPT_INV_VIA_IN)) {
dprintf("VIA in mismatch (%s vs %s).%s\n",
indev, arpinfo->iniface,
arpinfo->invflags&ARPT_INV_VIA_IN ?" (INV)":"");
return 0;
}
for (i = 0, ret = 0; i < IFNAMSIZ/sizeof(unsigned long); i++) {
ret |= (((const unsigned long *)outdev)[i]
^ ((const unsigned long *)arpinfo->outiface)[i])
& ((const unsigned long *)arpinfo->outiface_mask)[i];
}
if (FWINV(ret != 0, ARPT_INV_VIA_OUT)) {
dprintf("VIA out mismatch (%s vs %s).%s\n",
outdev, arpinfo->outiface,
arpinfo->invflags&ARPT_INV_VIA_OUT ?" (INV)":"");
return 0;
}
return 1;
}
static inline int arp_checkentry(const struct arpt_arp *arp)
{
if (arp->flags & ~ARPT_F_MASK) {
duprintf("Unknown flag bits set: %08X\n",
arp->flags & ~ARPT_F_MASK);
return 0;
}
if (arp->invflags & ~ARPT_INV_MASK) {
duprintf("Unknown invflag bits set: %08X\n",
arp->invflags & ~ARPT_INV_MASK);
return 0;
}
return 1;
}
static unsigned int arpt_error(struct sk_buff **pskb,
unsigned int hooknum,
const struct net_device *in,
const struct net_device *out,
const void *targinfo,
void *userinfo)
{
if (net_ratelimit())
printk("arp_tables: error: '%s'\n", (char *)targinfo);
return NF_DROP;
}
static inline struct arpt_entry *get_entry(void *base, unsigned int offset)
{
return (struct arpt_entry *)(base + offset);
}
unsigned int arpt_do_table(struct sk_buff **pskb,
unsigned int hook,
const struct net_device *in,
const struct net_device *out,
struct arpt_table *table,
void *userdata)
{
static const char nulldevname[IFNAMSIZ] = { 0 };
unsigned int verdict = NF_DROP;
struct arphdr *arp = (*pskb)->nh.arph;
int hotdrop = 0;
struct arpt_entry *e, *back;
const char *indev, *outdev;
void *table_base;
indev = in ? in->name : nulldevname;
outdev = out ? out->name : nulldevname;
read_lock_bh(&table->lock);
table_base = (void *)table->private->entries
+ TABLE_OFFSET(table->private,
cpu_number_map(smp_processor_id()));
e = get_entry(table_base, table->private->hook_entry[hook]);
back = get_entry(table_base, table->private->underflow[hook]);
do {
if (arp_packet_match(arp, (*pskb)->dev, indev, outdev, &e->arp)) {
struct arpt_entry_target *t;
int hdr_len;
hdr_len = sizeof(*arp) + (2 * sizeof(struct in_addr)) +
(2 * (*pskb)->dev->addr_len);
ADD_COUNTER(e->counters, hdr_len, 1);
t = arpt_get_target(e);
/* Standard target? */
if (!t->u.kernel.target->target) {
int v;
v = ((struct arpt_standard_target *)t)->verdict;
if (v < 0) {
/* Pop from stack? */
if (v != ARPT_RETURN) {
verdict = (unsigned)(-v) - 1;
break;
}
e = back;
back = get_entry(table_base,
back->comefrom);
continue;
}
if (table_base + v
!= (void *)e + e->next_offset) {
/* Save old back ptr in next entry */
struct arpt_entry *next
= (void *)e + e->next_offset;
next->comefrom =
(void *)back - table_base;
/* set back pointer to next entry */
back = next;
}
e = get_entry(table_base, v);
} else {
/* Targets which reenter must return
* abs. verdicts
*/
verdict = t->u.kernel.target->target(pskb,
hook,
in, out,
t->data,
userdata);
/* Target might have changed stuff. */
arp = (*pskb)->nh.arph;
if (verdict == ARPT_CONTINUE)
e = (void *)e + e->next_offset;
else
/* Verdict */
break;
}
} else {
e = (void *)e + e->next_offset;
}
} while (!hotdrop);
read_unlock_bh(&table->lock);
if (hotdrop)
return NF_DROP;
else
return verdict;
}
static inline void *find_inlist_lock_noload(struct list_head *head,
const char *name,
int *error,
struct semaphore *mutex)
{
void *ret;
*error = down_interruptible(mutex);
if (*error != 0)
return NULL;
ret = list_named_find(head, name);
if (!ret) {
*error = -ENOENT;
up(mutex);
}
return ret;
}
#ifndef CONFIG_KMOD
#define find_inlist_lock(h,n,p,e,m) find_inlist_lock_noload((h),(n),(e),(m))
#else
static void *
find_inlist_lock(struct list_head *head,
const char *name,
const char *prefix,
int *error,
struct semaphore *mutex)
{
void *ret;
ret = find_inlist_lock_noload(head, name, error, mutex);
if (!ret) {
char modulename[ARPT_FUNCTION_MAXNAMELEN + strlen(prefix) + 1];
strcpy(modulename, prefix);
strcat(modulename, name);
duprintf("find_inlist: loading `%s'.\n", modulename);
request_module(modulename);
ret = find_inlist_lock_noload(head, name, error, mutex);
}
return ret;
}
#endif
static inline struct arpt_table *find_table_lock(const char *name, int *error, struct semaphore *mutex)
{
return find_inlist_lock(&arpt_tables, name, "arptable_", error, mutex);
}
static inline struct arpt_target *find_target_lock(const char *name, int *error, struct semaphore *mutex)
{
return find_inlist_lock(&arpt_target, name, "arpt_", error, mutex);
}
/* All zeroes == unconditional rule. */
static inline int unconditional(const struct arpt_arp *arp)
{
unsigned int i;
for (i = 0; i < sizeof(*arp)/sizeof(__u32); i++)
if (((__u32 *)arp)[i])
return 0;
return 1;
}
/* Figures out from what hook each rule can be called: returns 0 if
* there are loops. Puts hook bitmask in comefrom.
*/
static int mark_source_chains(struct arpt_table_info *newinfo, unsigned int valid_hooks)
{
unsigned int hook;
/* No recursion; use packet counter to save back ptrs (reset
* to 0 as we leave), and comefrom to save source hook bitmask.
*/
for (hook = 0; hook < NF_ARP_NUMHOOKS; hook++) {
unsigned int pos = newinfo->hook_entry[hook];
struct arpt_entry *e
= (struct arpt_entry *)(newinfo->entries + pos);
if (!(valid_hooks & (1 << hook)))
continue;
/* Set initial back pointer. */
e->counters.pcnt = pos;
for (;;) {
struct arpt_standard_target *t
= (void *)arpt_get_target(e);
if (e->comefrom & (1 << NF_ARP_NUMHOOKS)) {
printk("arptables: loop hook %u pos %u %08X.\n",
hook, pos, e->comefrom);
return 0;
}
e->comefrom
|= ((1 << hook) | (1 << NF_ARP_NUMHOOKS));
/* Unconditional return/END. */
if (e->target_offset == sizeof(struct arpt_entry)
&& (strcmp(t->target.u.user.name,
ARPT_STANDARD_TARGET) == 0)
&& t->verdict < 0
&& unconditional(&e->arp)) {
unsigned int oldpos, size;
/* Return: backtrack through the last
* big jump.
*/
do {
e->comefrom ^= (1<<NF_ARP_NUMHOOKS);
oldpos = pos;
pos = e->counters.pcnt;
e->counters.pcnt = 0;
/* We're at the start. */
if (pos == oldpos)
goto next;
e = (struct arpt_entry *)
(newinfo->entries + pos);
} while (oldpos == pos + e->next_offset);
/* Move along one */
size = e->next_offset;
e = (struct arpt_entry *)
(newinfo->entries + pos + size);
e->counters.pcnt = pos;
pos += size;
} else {
int newpos = t->verdict;
if (strcmp(t->target.u.user.name,
ARPT_STANDARD_TARGET) == 0
&& newpos >= 0) {
/* This a jump; chase it. */
duprintf("Jump rule %u -> %u\n",
pos, newpos);
} else {
/* ... this is a fallthru */
newpos = pos + e->next_offset;
}
e = (struct arpt_entry *)
(newinfo->entries + newpos);
e->counters.pcnt = pos;
pos = newpos;
}
}
next:
duprintf("Finished chain %u\n", hook);
}
return 1;
}
static inline int standard_check(const struct arpt_entry_target *t,
unsigned int max_offset)
{
struct arpt_standard_target *targ = (void *)t;
/* Check standard info. */
if (t->u.target_size
!= ARPT_ALIGN(sizeof(struct arpt_standard_target))) {
duprintf("arpt_standard_check: target size %u != %Zu\n",
t->u.target_size,
ARPT_ALIGN(sizeof(struct arpt_standard_target)));
return 0;
}
if (targ->verdict >= 0
&& targ->verdict > max_offset - sizeof(struct arpt_entry)) {
duprintf("arpt_standard_check: bad verdict (%i)\n",
targ->verdict);
return 0;
}
if (targ->verdict < -NF_MAX_VERDICT - 1) {
duprintf("arpt_standard_check: bad negative verdict (%i)\n",
targ->verdict);
return 0;
}
return 1;
}
static struct arpt_target arpt_standard_target;
static inline int check_entry(struct arpt_entry *e, const char *name, unsigned int size,
unsigned int *i)
{
struct arpt_entry_target *t;
struct arpt_target *target;
int ret;
if (!arp_checkentry(&e->arp)) {
duprintf("arp_tables: arp check failed %p %s.\n", e, name);
return -EINVAL;
}
t = arpt_get_target(e);
target = find_target_lock(t->u.user.name, &ret, &arpt_mutex);
if (!target) {
duprintf("check_entry: `%s' not found\n", t->u.user.name);
goto out;
}
if (target->me)
__MOD_INC_USE_COUNT(target->me);
t->u.kernel.target = target;
up(&arpt_mutex);
if (t->u.kernel.target == &arpt_standard_target) {
if (!standard_check(t, size)) {
ret = -EINVAL;
goto out;
}
} else if (t->u.kernel.target->checkentry
&& !t->u.kernel.target->checkentry(name, e, t->data,
t->u.target_size
- sizeof(*t),
e->comefrom)) {
if (t->u.kernel.target->me)
__MOD_DEC_USE_COUNT(t->u.kernel.target->me);
duprintf("arp_tables: check failed for `%s'.\n",
t->u.kernel.target->name);
ret = -EINVAL;
goto out;
}
(*i)++;
return 0;
out:
return ret;
}
static inline int check_entry_size_and_hooks(struct arpt_entry *e,
struct arpt_table_info *newinfo,
unsigned char *base,
unsigned char *limit,
const unsigned int *hook_entries,
const unsigned int *underflows,
unsigned int *i)
{
unsigned int h;
if ((unsigned long)e % __alignof__(struct arpt_entry) != 0
|| (unsigned char *)e + sizeof(struct arpt_entry) >= limit) {
duprintf("Bad offset %p\n", e);
return -EINVAL;
}
if (e->next_offset
< sizeof(struct arpt_entry) + sizeof(struct arpt_entry_target)) {
duprintf("checking: element %p size %u\n",
e, e->next_offset);
return -EINVAL;
}
/* Check hooks & underflows */
for (h = 0; h < NF_ARP_NUMHOOKS; h++) {
if ((unsigned char *)e - base == hook_entries[h])
newinfo->hook_entry[h] = hook_entries[h];
if ((unsigned char *)e - base == underflows[h])
newinfo->underflow[h] = underflows[h];
}
/* FIXME: underflows must be unconditional, standard verdicts
< 0 (not ARPT_RETURN). --RR */
/* Clear counters and comefrom */
e->counters = ((struct arpt_counters) { 0, 0 });
e->comefrom = 0;
(*i)++;
return 0;
}
static inline int cleanup_entry(struct arpt_entry *e, unsigned int *i)
{
struct arpt_entry_target *t;
if (i && (*i)-- == 0)
return 1;
t = arpt_get_target(e);
if (t->u.kernel.target->destroy)
t->u.kernel.target->destroy(t->data,
t->u.target_size - sizeof(*t));
if (t->u.kernel.target->me)
__MOD_DEC_USE_COUNT(t->u.kernel.target->me);
return 0;
}
/* Checks and translates the user-supplied table segment (held in
* newinfo).
*/
static int translate_table(const char *name,
unsigned int valid_hooks,
struct arpt_table_info *newinfo,
unsigned int size,
unsigned int number,
const unsigned int *hook_entries,
const unsigned int *underflows)
{
unsigned int i;
int ret;
newinfo->size = size;
newinfo->number = number;
/* Init all hooks to impossible value. */
for (i = 0; i < NF_ARP_NUMHOOKS; i++) {
newinfo->hook_entry[i] = 0xFFFFFFFF;
newinfo->underflow[i] = 0xFFFFFFFF;
}
duprintf("translate_table: size %u\n", newinfo->size);
i = 0;
/* Walk through entries, checking offsets. */
ret = ARPT_ENTRY_ITERATE(newinfo->entries, newinfo->size,
check_entry_size_and_hooks,
newinfo,
newinfo->entries,
newinfo->entries + size,
hook_entries, underflows, &i);
duprintf("translate_table: ARPT_ENTRY_ITERATE gives %d\n", ret);
if (ret != 0)
return ret;
if (i != number) {
duprintf("translate_table: %u not %u entries\n",
i, number);
return -EINVAL;
}
/* Check hooks all assigned */
for (i = 0; i < NF_ARP_NUMHOOKS; i++) {
/* Only hooks which are valid */
if (!(valid_hooks & (1 << i)))
continue;
if (newinfo->hook_entry[i] == 0xFFFFFFFF) {
duprintf("Invalid hook entry %u %u\n",
i, hook_entries[i]);
return -EINVAL;
}
if (newinfo->underflow[i] == 0xFFFFFFFF) {
duprintf("Invalid underflow %u %u\n",
i, underflows[i]);
return -EINVAL;
}
}
if (!mark_source_chains(newinfo, valid_hooks)) {
duprintf("Looping hook\n");
return -ELOOP;
}
/* Finally, each sanity check must pass */
i = 0;
ret = ARPT_ENTRY_ITERATE(newinfo->entries, newinfo->size,
check_entry, name, size, &i);
if (ret != 0) {
ARPT_ENTRY_ITERATE(newinfo->entries, newinfo->size,
cleanup_entry, &i);
return ret;
}
/* And one copy for every other CPU */
for (i = 1; i < smp_num_cpus; i++) {
memcpy(newinfo->entries + SMP_ALIGN(newinfo->size)*i,
newinfo->entries,
SMP_ALIGN(newinfo->size));
}
return ret;
}
static struct arpt_table_info *replace_table(struct arpt_table *table,
unsigned int num_counters,
struct arpt_table_info *newinfo,
int *error)
{
struct arpt_table_info *oldinfo;
/* Do the substitution. */
write_lock_bh(&table->lock);
/* Check inside lock: is the old number correct? */
if (num_counters != table->private->number) {
duprintf("num_counters != table->private->number (%u/%u)\n",
num_counters, table->private->number);
write_unlock_bh(&table->lock);
*error = -EAGAIN;
return NULL;
}
oldinfo = table->private;
table->private = newinfo;
newinfo->initial_entries = oldinfo->initial_entries;
write_unlock_bh(&table->lock);
return oldinfo;
}
/* Gets counters. */
static inline int add_entry_to_counter(const struct arpt_entry *e,
struct arpt_counters total[],
unsigned int *i)
{
ADD_COUNTER(total[*i], e->counters.bcnt, e->counters.pcnt);
(*i)++;
return 0;
}
static void get_counters(const struct arpt_table_info *t,
struct arpt_counters counters[])
{
unsigned int cpu;
unsigned int i;
for (cpu = 0; cpu < smp_num_cpus; cpu++) {
i = 0;
ARPT_ENTRY_ITERATE(t->entries + TABLE_OFFSET(t, cpu),
t->size,
add_entry_to_counter,
counters,
&i);
}
}
static int copy_entries_to_user(unsigned int total_size,
struct arpt_table *table,
void *userptr)
{
unsigned int off, num, countersize;
struct arpt_entry *e;
struct arpt_counters *counters;
int ret = 0;
/* We need atomic snapshot of counters: rest doesn't change
* (other than comefrom, which userspace doesn't care
* about).
*/
countersize = sizeof(struct arpt_counters) * table->private->number;
counters = vmalloc(countersize);
if (counters == NULL)
return -ENOMEM;
/* First, sum counters... */
memset(counters, 0, countersize);
write_lock_bh(&table->lock);
get_counters(table->private, counters);
write_unlock_bh(&table->lock);
/* ... then copy entire thing from CPU 0... */
if (copy_to_user(userptr, table->private->entries, total_size) != 0) {
ret = -EFAULT;
goto free_counters;
}
/* FIXME: use iterator macros --RR */
/* ... then go back and fix counters and names */
for (off = 0, num = 0; off < total_size; off += e->next_offset, num++){
struct arpt_entry_target *t;
e = (struct arpt_entry *)(table->private->entries + off);
if (copy_to_user(userptr + off
+ offsetof(struct arpt_entry, counters),
&counters[num],
sizeof(counters[num])) != 0) {
ret = -EFAULT;
goto free_counters;
}
t = arpt_get_target(e);
if (copy_to_user(userptr + off + e->target_offset
+ offsetof(struct arpt_entry_target,
u.user.name),
t->u.kernel.target->name,
strlen(t->u.kernel.target->name)+1) != 0) {
ret = -EFAULT;
goto free_counters;
}
}
free_counters:
vfree(counters);
return ret;
}
static int get_entries(const struct arpt_get_entries *entries,
struct arpt_get_entries *uptr)
{
int ret;
struct arpt_table *t;
t = find_table_lock(entries->name, &ret, &arpt_mutex);
if (t) {
duprintf("t->private->number = %u\n",
t->private->number);
if (entries->size == t->private->size)
ret = copy_entries_to_user(t->private->size,
t, uptr->entrytable);
else {
duprintf("get_entries: I've got %u not %u!\n",
t->private->size,
entries->size);
ret = -EINVAL;
}
up(&arpt_mutex);
} else
duprintf("get_entries: Can't find %s!\n",
entries->name);
return ret;
}
static int do_replace(void *user, unsigned int len)
{
int ret;
struct arpt_replace tmp;
struct arpt_table *t;
struct arpt_table_info *newinfo, *oldinfo;
struct arpt_counters *counters;
if (copy_from_user(&tmp, user, sizeof(tmp)) != 0)
return -EFAULT;
/* Hack: Causes ipchains to give correct error msg --RR */
if (len != sizeof(tmp) + tmp.size)
return -ENOPROTOOPT;
/* Pedantry: prevent them from hitting BUG() in vmalloc.c --RR */
if ((SMP_ALIGN(tmp.size) >> PAGE_SHIFT) + 2 > num_physpages)
return -ENOMEM;
newinfo = vmalloc(sizeof(struct arpt_table_info)
+ SMP_ALIGN(tmp.size) * smp_num_cpus);
if (!newinfo)
return -ENOMEM;
if (copy_from_user(newinfo->entries, user + sizeof(tmp),
tmp.size) != 0) {
ret = -EFAULT;
goto free_newinfo;
}
counters = vmalloc(tmp.num_counters * sizeof(struct arpt_counters));
if (!counters) {
ret = -ENOMEM;
goto free_newinfo;
}
memset(counters, 0, tmp.num_counters * sizeof(struct arpt_counters));
ret = translate_table(tmp.name, tmp.valid_hooks,
newinfo, tmp.size, tmp.num_entries,
tmp.hook_entry, tmp.underflow);
if (ret != 0)
goto free_newinfo_counters;
duprintf("arp_tables: Translated table\n");
t = find_table_lock(tmp.name, &ret, &arpt_mutex);
if (!t)
goto free_newinfo_counters_untrans;
/* You lied! */
if (tmp.valid_hooks != t->valid_hooks) {
duprintf("Valid hook crap: %08X vs %08X\n",
tmp.valid_hooks, t->valid_hooks);
ret = -EINVAL;
goto free_newinfo_counters_untrans_unlock;
}
oldinfo = replace_table(t, tmp.num_counters, newinfo, &ret);
if (!oldinfo)
goto free_newinfo_counters_untrans_unlock;
/* Update module usage count based on number of rules */
duprintf("do_replace: oldnum=%u, initnum=%u, newnum=%u\n",
oldinfo->number, oldinfo->initial_entries, newinfo->number);
if (t->me && (oldinfo->number <= oldinfo->initial_entries) &&
(newinfo->number > oldinfo->initial_entries))
__MOD_INC_USE_COUNT(t->me);
else if (t->me && (oldinfo->number > oldinfo->initial_entries) &&
(newinfo->number <= oldinfo->initial_entries))
__MOD_DEC_USE_COUNT(t->me);
/* Get the old counters. */
get_counters(oldinfo, counters);
/* Decrease module usage counts and free resource */
ARPT_ENTRY_ITERATE(oldinfo->entries, oldinfo->size, cleanup_entry,NULL);
vfree(oldinfo);
/* Silent error: too late now. */
copy_to_user(tmp.counters, counters,
sizeof(struct arpt_counters) * tmp.num_counters);
vfree(counters);
up(&arpt_mutex);
return 0;
free_newinfo_counters_untrans_unlock:
up(&arpt_mutex);
free_newinfo_counters_untrans:
ARPT_ENTRY_ITERATE(newinfo->entries, newinfo->size, cleanup_entry, NULL);
free_newinfo_counters:
vfree(counters);
free_newinfo:
vfree(newinfo);
return ret;
}
/* We're lazy, and add to the first CPU; overflow works its fey magic
* and everything is OK.
*/
static inline int add_counter_to_entry(struct arpt_entry *e,
const struct arpt_counters addme[],
unsigned int *i)
{
ADD_COUNTER(e->counters, addme[*i].bcnt, addme[*i].pcnt);
(*i)++;
return 0;
}
static int do_add_counters(void *user, unsigned int len)
{
unsigned int i;
struct arpt_counters_info tmp, *paddc;
struct arpt_table *t;
int ret;
if (copy_from_user(&tmp, user, sizeof(tmp)) != 0)
return -EFAULT;
if (len != sizeof(tmp) + tmp.num_counters*sizeof(struct arpt_counters))
return -EINVAL;
paddc = vmalloc(len);
if (!paddc)
return -ENOMEM;
if (copy_from_user(paddc, user, len) != 0) {
ret = -EFAULT;
goto free;
}
t = find_table_lock(tmp.name, &ret, &arpt_mutex);
if (!t)
goto free;
write_lock_bh(&t->lock);
if (t->private->number != paddc->num_counters) {
ret = -EINVAL;
goto unlock_up_free;
}
i = 0;
ARPT_ENTRY_ITERATE(t->private->entries,
t->private->size,
add_counter_to_entry,
paddc->counters,
&i);
unlock_up_free:
write_unlock_bh(&t->lock);
up(&arpt_mutex);
free:
vfree(paddc);
return ret;
}
static int do_arpt_set_ctl(struct sock *sk, int cmd, void *user, unsigned int len)
{
int ret;
if (!capable(CAP_NET_ADMIN))
return -EPERM;
switch (cmd) {
case ARPT_SO_SET_REPLACE:
ret = do_replace(user, len);
break;
case ARPT_SO_SET_ADD_COUNTERS:
ret = do_add_counters(user, len);
break;
default:
duprintf("do_arpt_set_ctl: unknown request %i\n", cmd);
ret = -EINVAL;
}
return ret;
}
static int do_arpt_get_ctl(struct sock *sk, int cmd, void *user, int *len)
{
int ret;
if (!capable(CAP_NET_ADMIN))
return -EPERM;
switch (cmd) {
case ARPT_SO_GET_INFO: {
char name[ARPT_TABLE_MAXNAMELEN];
struct arpt_table *t;
if (*len != sizeof(struct arpt_getinfo)) {
duprintf("length %u != %Zu\n", *len,
sizeof(struct arpt_getinfo));
ret = -EINVAL;
break;
}
if (copy_from_user(name, user, sizeof(name)) != 0) {
ret = -EFAULT;
break;
}
name[ARPT_TABLE_MAXNAMELEN-1] = '\0';
t = find_table_lock(name, &ret, &arpt_mutex);
if (t) {
struct arpt_getinfo info;
info.valid_hooks = t->valid_hooks;
memcpy(info.hook_entry, t->private->hook_entry,
sizeof(info.hook_entry));
memcpy(info.underflow, t->private->underflow,
sizeof(info.underflow));
info.num_entries = t->private->number;
info.size = t->private->size;
strcpy(info.name, name);
if (copy_to_user(user, &info, *len) != 0)
ret = -EFAULT;
else
ret = 0;
up(&arpt_mutex);
}
}
break;
case ARPT_SO_GET_ENTRIES: {
struct arpt_get_entries get;
if (*len < sizeof(get)) {
duprintf("get_entries: %u < %Zu\n", *len, sizeof(get));
ret = -EINVAL;
} else if (copy_from_user(&get, user, sizeof(get)) != 0) {
ret = -EFAULT;
} else if (*len != sizeof(struct arpt_get_entries) + get.size) {
duprintf("get_entries: %u != %Zu\n", *len,
sizeof(struct arpt_get_entries) + get.size);
ret = -EINVAL;
} else
ret = get_entries(&get, user);
break;
}
default:
duprintf("do_arpt_get_ctl: unknown request %i\n", cmd);
ret = -EINVAL;
}
return ret;
}
/* Registration hooks for targets. */
int arpt_register_target(struct arpt_target *target)
{
int ret;
MOD_INC_USE_COUNT;
ret = down_interruptible(&arpt_mutex);
if (ret != 0) {
MOD_DEC_USE_COUNT;
return ret;
}
if (!list_named_insert(&arpt_target, target)) {
duprintf("arpt_register_target: `%s' already in list!\n",
target->name);
ret = -EINVAL;
MOD_DEC_USE_COUNT;
}
up(&arpt_mutex);
return ret;
}
void arpt_unregister_target(struct arpt_target *target)
{
down(&arpt_mutex);
LIST_DELETE(&arpt_target, target);
up(&arpt_mutex);
MOD_DEC_USE_COUNT;
}
int arpt_register_table(struct arpt_table *table)
{
int ret;
struct arpt_table_info *newinfo;
static struct arpt_table_info bootstrap
= { 0, 0, 0, { 0 }, { 0 }, { } };
MOD_INC_USE_COUNT;
newinfo = vmalloc(sizeof(struct arpt_table_info)
+ SMP_ALIGN(table->table->size) * smp_num_cpus);
if (!newinfo) {
ret = -ENOMEM;
MOD_DEC_USE_COUNT;
return ret;
}
memcpy(newinfo->entries, table->table->entries, table->table->size);
ret = translate_table(table->name, table->valid_hooks,
newinfo, table->table->size,
table->table->num_entries,
table->table->hook_entry,
table->table->underflow);
duprintf("arpt_register_table: translate table gives %d\n", ret);
if (ret != 0) {
vfree(newinfo);
MOD_DEC_USE_COUNT;
return ret;
}
ret = down_interruptible(&arpt_mutex);
if (ret != 0) {
vfree(newinfo);
MOD_DEC_USE_COUNT;
return ret;
}
/* Don't autoload: we'd eat our tail... */
if (list_named_find(&arpt_tables, table->name)) {
ret = -EEXIST;
goto free_unlock;
}
/* Simplifies replace_table code. */
table->private = &bootstrap;
if (!replace_table(table, 0, newinfo, &ret))
goto free_unlock;
duprintf("table->private->number = %u\n",
table->private->number);
/* save number of initial entries */
table->private->initial_entries = table->private->number;
table->lock = RW_LOCK_UNLOCKED;
list_prepend(&arpt_tables, table);
unlock:
up(&arpt_mutex);
return ret;
free_unlock:
vfree(newinfo);
MOD_DEC_USE_COUNT;
goto unlock;
}
void arpt_unregister_table(struct arpt_table *table)
{
down(&arpt_mutex);
LIST_DELETE(&arpt_tables, table);
up(&arpt_mutex);
/* Decrease module usage counts and free resources */
ARPT_ENTRY_ITERATE(table->private->entries, table->private->size,
cleanup_entry, NULL);
vfree(table->private);
MOD_DEC_USE_COUNT;
}
/* The built-in targets: standard (NULL) and error. */
static struct arpt_target arpt_standard_target
= { { NULL, NULL }, ARPT_STANDARD_TARGET, NULL, NULL, NULL };
static struct arpt_target arpt_error_target
= { { NULL, NULL }, ARPT_ERROR_TARGET, arpt_error, NULL, NULL };
static struct nf_sockopt_ops arpt_sockopts
= { { NULL, NULL }, PF_INET, ARPT_BASE_CTL, ARPT_SO_SET_MAX+1, do_arpt_set_ctl,
ARPT_BASE_CTL, ARPT_SO_GET_MAX+1, do_arpt_get_ctl, 0, NULL };
#ifdef CONFIG_PROC_FS
static inline int print_name(const struct arpt_table *t,
off_t start_offset, char *buffer, int length,
off_t *pos, unsigned int *count)
{
if ((*count)++ >= start_offset) {
unsigned int namelen;
namelen = sprintf(buffer + *pos, "%s\n", t->name);
if (*pos + namelen > length) {
/* Stop iterating */
return 1;
}
*pos += namelen;
}
return 0;
}
static int arpt_get_tables(char *buffer, char **start, off_t offset, int length)
{
off_t pos = 0;
unsigned int count = 0;
if (down_interruptible(&arpt_mutex) != 0)
return 0;
LIST_FIND(&arpt_tables, print_name, struct arpt_table *,
offset, buffer, length, &pos, &count);
up(&arpt_mutex);
/* `start' hack - see fs/proc/generic.c line ~105 */
*start=(char *)((unsigned long)count-offset);
return pos;
}
#endif /*CONFIG_PROC_FS*/
static int __init init(void)
{
int ret;
/* Noone else will be downing sem now, so we won't sleep */
down(&arpt_mutex);
list_append(&arpt_target, &arpt_standard_target);
list_append(&arpt_target, &arpt_error_target);
up(&arpt_mutex);
/* Register setsockopt */
ret = nf_register_sockopt(&arpt_sockopts);
if (ret < 0) {
duprintf("Unable to register sockopts.\n");
return ret;
}
#ifdef CONFIG_PROC_FS
{
struct proc_dir_entry *proc;
proc = proc_net_create("arp_tables_names", 0, arpt_get_tables);
if (!proc) {
nf_unregister_sockopt(&arpt_sockopts);
return -ENOMEM;
}
proc->owner = THIS_MODULE;
}
#endif
printk("arp_tables: (C) 2002 David S. Miller\n");
return 0;
}
static void __exit fini(void)
{
nf_unregister_sockopt(&arpt_sockopts);
#ifdef CONFIG_PROC_FS
proc_net_remove("arp_tables_names");
#endif
}
EXPORT_SYMBOL(arpt_register_table);
EXPORT_SYMBOL(arpt_unregister_table);
EXPORT_SYMBOL(arpt_do_table);
EXPORT_SYMBOL(arpt_register_target);
EXPORT_SYMBOL(arpt_unregister_target);
module_init(init);
module_exit(fini);
MODULE_LICENSE("GPL");
net/ipv4/netfilter/arptable_filter.c 0 → 100644
View file @ 4ce937cb
/*
* Filtering ARP tables module.
*
* Copyright (C) 2002 David S. Miller (davem@redhat.com)
*
*/
#include <linux/module.h>
#include <linux/netfilter_arp/arp_tables.h>
#define FILTER_VALID_HOOKS ((1 << NF_ARP_IN) | (1 << NF_ARP_OUT))
/* Standard entry. */
struct arpt_standard
{
struct arpt_entry entry;
struct arpt_standard_target target;
};
struct arpt_error_target
{
struct arpt_entry_target target;
char errorname[ARPT_FUNCTION_MAXNAMELEN];
};
struct arpt_error
{
struct arpt_entry entry;
struct arpt_error_target target;
};
static struct
{
struct arpt_replace repl;
struct arpt_standard entries[2];
struct arpt_error term;
} initial_table __initdata
= { { "filter", FILTER_VALID_HOOKS, 3,
sizeof(struct arpt_standard) * 2 + sizeof(struct arpt_error),
{ [NF_ARP_IN] 0,
[NF_ARP_OUT] sizeof(struct arpt_standard) },
{ [NF_ARP_IN] 0,
[NF_ARP_OUT] sizeof(struct arpt_standard), },
0, NULL, { } },
{
/* ARP_IN */
{
{
{
{ 0 }, { 0 }, { 0 }, { 0 },
0, 0,
{ { 0, }, { 0, } },
{ { 0, }, { 0, } },
0, 0,
0, 0,
0, 0,
"", "", { 0 }, { 0 },
0, 0
},
sizeof(struct arpt_entry),
sizeof(struct arpt_standard),
0,
{ 0, 0 }, { } },
{ { { { ARPT_ALIGN(sizeof(struct arpt_standard_target)), "" } }, { } },
-NF_ACCEPT - 1 }
},
/* ARP_OUT */
{
{
{
{ 0 }, { 0 }, { 0 }, { 0 },
0, 0,
{ { 0, }, { 0, } },
{ { 0, }, { 0, } },
0, 0,
0, 0,
0, 0,
"", "", { 0 }, { 0 },
0, 0
},
sizeof(struct arpt_entry),
sizeof(struct arpt_standard),
0,
{ 0, 0 }, { } },
{ { { { ARPT_ALIGN(sizeof(struct arpt_standard_target)), "" } }, { } },
-NF_ACCEPT - 1 }
}
},
/* ERROR */
{
{
{
{ 0 }, { 0 }, { 0 }, { 0 },
0, 0,
{ { 0, }, { 0, } },
{ { 0, }, { 0, } },
0, 0,
0, 0,
0, 0,
"", "", { 0 }, { 0 },
0, 0
},
sizeof(struct arpt_entry),
sizeof(struct arpt_error),
0,
{ 0, 0 }, { } },
{ { { { ARPT_ALIGN(sizeof(struct arpt_error_target)), ARPT_ERROR_TARGET } },
{ } },
"ERROR"
}
}
};
static struct arpt_table packet_filter
= { { NULL, NULL }, "filter", &initial_table.repl,
FILTER_VALID_HOOKS, RW_LOCK_UNLOCKED, NULL, THIS_MODULE };
/* The work comes in here from netfilter.c */
static unsigned int arpt_hook(unsigned int hook,
struct sk_buff **pskb,
const struct net_device *in,
const struct net_device *out,
int (*okfn)(struct sk_buff *))
{
return arpt_do_table(pskb, hook, in, out, &packet_filter, NULL);
}
static struct nf_hook_ops arpt_ops[]
= { { { NULL, NULL }, arpt_hook, NF_ARP, NF_ARP_IN, 0 },
{ { NULL, NULL }, arpt_hook, NF_ARP, NF_ARP_OUT, 0 }
};
static int __init init(void)
{
int ret;
/* Register table */
ret = arpt_register_table(&packet_filter);
if (ret < 0)
return ret;
/* Register hooks */
ret = nf_register_hook(&arpt_ops[0]);
if (ret < 0)
goto cleanup_table;
ret = nf_register_hook(&arpt_ops[1]);
if (ret < 0)
goto cleanup_hook0;
return ret;
cleanup_hook0:
nf_unregister_hook(&arpt_ops[0]);
cleanup_table:
arpt_unregister_table(&packet_filter);
return ret;
}
static void __exit fini(void)
{
unsigned int i;
for (i = 0; i < sizeof(arpt_ops)/sizeof(struct nf_hook_ops); i++)
nf_unregister_hook(&arpt_ops[i]);
arpt_unregister_table(&packet_filter);
}
module_init(init);
module_exit(fini);
MODULE_LICENSE("GPL");
net/ipv4/netfilter/ip_conntrack_standalone.c
View file @ 4ce937cb
......@@ -15,6 +15,7 @@
#include <linux/skbuff.h>
#include <linux/proc_fs.h>
#include <linux/version.h>
#include <linux/brlock.h>
#include <net/checksum.h>
#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_conntrack_lock)
......@@ -35,6 +36,11 @@
struct module *ip_conntrack_module = THIS_MODULE;
MODULE_LICENSE("GPL");
static int kill_proto(const struct ip_conntrack *i, void *data)
{
return (i->tuplehash[IP_CT_DIR_ORIGINAL].dst.protonum ==
*((u_int8_t *) data));
}
static unsigned int
print_tuple(char *buffer, const struct ip_conntrack_tuple *tuple,
......@@ -304,12 +310,24 @@ int ip_conntrack_protocol_register(struct ip_conntrack_protocol *proto)
return ret;
}
/* FIXME: Implement this --RR */
#if 0
void ip_conntrack_protocol_unregister(struct ip_conntrack_protocol *proto)
{
WRITE_LOCK(&ip_conntrack_lock);
/* find_proto() returns proto_generic in case there is no protocol
* helper. So this should be enough - HW */
LIST_DELETE(&protocol_list, proto);
WRITE_UNLOCK(&ip_conntrack_lock);
/* Somebody could be still looking at the proto in bh. */
br_write_lock_bh(BR_NETPROTO_LOCK);
br_write_unlock_bh(BR_NETPROTO_LOCK);
/* Remove all contrack entries for this protocol */
ip_ct_selective_cleanup(kill_proto, &proto->proto);
MOD_DEC_USE_COUNT;
}
#endif
static int __init init(void)
{
......@@ -325,6 +343,7 @@ module_init(init);
module_exit(fini);
EXPORT_SYMBOL(ip_conntrack_protocol_register);
EXPORT_SYMBOL(ip_conntrack_protocol_unregister);
EXPORT_SYMBOL(invert_tuplepr);
EXPORT_SYMBOL(ip_conntrack_alter_reply);
EXPORT_SYMBOL(ip_conntrack_destroyed);
......@@ -335,6 +354,7 @@ EXPORT_SYMBOL(ip_conntrack_helper_unregister);
EXPORT_SYMBOL(ip_ct_selective_cleanup);
EXPORT_SYMBOL(ip_ct_refresh);
EXPORT_SYMBOL(ip_conntrack_expect_related);
EXPORT_SYMBOL(ip_conntrack_unexpect_related);
EXPORT_SYMBOL(ip_conntrack_tuple_taken);
EXPORT_SYMBOL(ip_ct_gather_frags);
EXPORT_SYMBOL(ip_conntrack_htable_size);
net/ipv4/netfilter/ip_nat_core.c
View file @ 4ce937cb
......@@ -314,6 +314,7 @@ find_best_ips_proto(struct ip_conntrack_tuple *tuple,
* do_extra_mangle last time. */
*other_ipp = saved_ip;
#ifdef CONFIG_IP_NF_NAT_LOCAL
if (hooknum == NF_IP_LOCAL_OUT
&& *var_ipp != orig_dstip
&& !do_extra_mangle(*var_ipp, other_ipp)) {
......@@ -324,6 +325,7 @@ find_best_ips_proto(struct ip_conntrack_tuple *tuple,
* anyway. */
continue;
}
#endif
/* Count how many others map onto this. */
score = count_maps(tuple->src.ip, tuple->dst.ip,
......@@ -367,11 +369,13 @@ find_best_ips_proto_fast(struct ip_conntrack_tuple *tuple,
else {
/* Only do extra mangle when required (breaks
socket binding) */
#ifdef CONFIG_IP_NF_NAT_LOCAL
if (tuple->dst.ip != mr->range[0].min_ip
&& hooknum == NF_IP_LOCAL_OUT
&& !do_extra_mangle(mr->range[0].min_ip,
&tuple->src.ip))
return NULL;
#endif
tuple->dst.ip = mr->range[0].min_ip;
}
}
......@@ -494,7 +498,10 @@ helper_cmp(const struct ip_nat_helper *helper,
static unsigned int opposite_hook[NF_IP_NUMHOOKS]
= { [NF_IP_PRE_ROUTING] = NF_IP_POST_ROUTING,
[NF_IP_POST_ROUTING] = NF_IP_PRE_ROUTING,
[NF_IP_LOCAL_OUT] = NF_IP_POST_ROUTING
#ifdef CONFIG_IP_NF_NAT_LOCAL
[NF_IP_LOCAL_OUT] = NF_IP_LOCAL_IN,
[NF_IP_LOCAL_IN] = NF_IP_LOCAL_OUT,
#endif
};
unsigned int
......
net/ipv4/netfilter/ip_nat_rule.c
View file @ 4ce937cb
......@@ -140,8 +140,12 @@ static unsigned int ipt_dnat_target(struct sk_buff **pskb,
struct ip_conntrack *ct;
enum ip_conntrack_info ctinfo;
#ifdef CONFIG_IP_NF_NAT_LOCAL
IP_NF_ASSERT(hooknum == NF_IP_PRE_ROUTING
|| hooknum == NF_IP_LOCAL_OUT);
#else
IP_NF_ASSERT(hooknum == NF_IP_PRE_ROUTING);
#endif
ct = ip_conntrack_get(*pskb, &ctinfo);
......@@ -210,7 +214,7 @@ static int ipt_dnat_checkentry(const char *tablename,
/* Only allow these for NAT. */
if (strcmp(tablename, "nat") != 0) {
DEBUGP("SNAT: wrong table %s\n", tablename);
DEBUGP("DNAT: wrong table %s\n", tablename);
return 0;
}
......@@ -218,6 +222,14 @@ static int ipt_dnat_checkentry(const char *tablename,
DEBUGP("DNAT: hook mask 0x%x bad\n", hook_mask);
return 0;
}
#ifndef CONFIG_IP_NF_NAT_LOCAL
if (hook_mask & (1 << NF_IP_LOCAL_OUT)) {
DEBUGP("DNAT: CONFIG_IP_NF_NAT_LOCAL not enabled\n");
return 0;
}
#endif
return 1;
}
......
net/ipv4/netfilter/ip_nat_standalone.c
View file @ 4ce937cb
......@@ -42,7 +42,8 @@
#define HOOKNAME(hooknum) ((hooknum) == NF_IP_POST_ROUTING ? "POST_ROUTING" \
: ((hooknum) == NF_IP_PRE_ROUTING ? "PRE_ROUTING" \
: ((hooknum) == NF_IP_LOCAL_OUT ? "LOCAL_OUT" \
: "*ERROR*")))
: ((hooknum) == NF_IP_LOCAL_IN ? "LOCAL_IN" \
: "*ERROR*")))
static unsigned int
ip_nat_fn(unsigned int hooknum,
......@@ -95,6 +96,12 @@ ip_nat_fn(unsigned int hooknum,
}
/* Fall thru... (Only ICMPs can be IP_CT_IS_REPLY) */
case IP_CT_NEW:
#ifdef CONFIG_IP_NF_NAT_LOCAL
/* LOCAL_IN hook doesn't have a chain and thus doesn't care
* about new packets -HW */
if (hooknum == NF_IP_LOCAL_IN)
return NF_ACCEPT;
#endif
info = &ct->nat.info;
WRITE_LOCK(&ip_nat_lock);
......@@ -205,6 +212,11 @@ static struct nf_hook_ops ip_nat_out_ops
static struct nf_hook_ops ip_nat_local_out_ops
= { { NULL, NULL }, ip_nat_local_fn, PF_INET, NF_IP_LOCAL_OUT, NF_IP_PRI_NAT_DST };
#ifdef CONFIG_IP_NF_NAT_LOCAL
static struct nf_hook_ops ip_nat_local_in_ops
= { { NULL, NULL }, ip_nat_fn, PF_INET, NF_IP_LOCAL_IN, NF_IP_PRI_NAT_SRC };
#endif
/* Protocol registration. */
int ip_nat_protocol_register(struct ip_nat_protocol *proto)
{
......@@ -273,6 +285,13 @@ static int init_or_cleanup(int init)
printk("ip_nat_init: can't register local out hook.\n");
goto cleanup_outops;
}
#ifdef CONFIG_IP_NF_NAT_LOCAL
ret = nf_register_hook(&ip_nat_local_in_ops);
if (ret < 0) {
printk("ip_nat_init: can't register local in hook.\n");
goto cleanup_localoutops;
}
#endif
if (ip_conntrack_module)
__MOD_INC_USE_COUNT(ip_conntrack_module);
return ret;
......@@ -280,6 +299,10 @@ static int init_or_cleanup(int init)
cleanup:
if (ip_conntrack_module)
__MOD_DEC_USE_COUNT(ip_conntrack_module);
#ifdef CONFIG_IP_NF_NAT_LOCAL
nf_unregister_hook(&ip_nat_local_in_ops);
cleanup_localoutops:
#endif
nf_unregister_hook(&ip_nat_local_out_ops);
cleanup_outops:
nf_unregister_hook(&ip_nat_out_ops);
......
net/netsyms.c
View file @ 4ce937cb
......@@ -490,6 +490,7 @@ EXPORT_SYMBOL(__kfree_skb);
EXPORT_SYMBOL(skb_clone);
EXPORT_SYMBOL(skb_copy);
EXPORT_SYMBOL(netif_rx);
EXPORT_SYMBOL(netif_receive_skb);
EXPORT_SYMBOL(dev_add_pack);
EXPORT_SYMBOL(dev_remove_pack);
EXPORT_SYMBOL(dev_get);
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7