pppoe_flush_dev() kicks all sockets bound to a device that is going down.
In doing so, locks must be taken in the right order consistently (sock lock,
followed by the pppoe_hash_lock).  However, the scan process is based on
us holding the sock lock.  So, when something is found in the scan we must
release the lock we're holding and grab the sock lock.

This patch fixes race conditions between this code and pppoe_release(),
both of which perform similar functions but would naturally prefer to grab
locks in opposing orders.  Both code paths are now going after these locks
in a consistent manner.

pppoe_hash_lock protects the contents of the "pppox_sock" objects that reside
inside the hash.  Thus, NULL'ing out the pppoe_dev field should be done
under the protection of this lock.
Signed-off-by: NMichal Ostrowski <mostrows@earthlink.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

below you find a patch that fixes a memory leak when a PPPoE socket is
release()d after it has been connect()ed, but before the PPPIOCGCHAN ioctl
ever has been called on it.

This is somewhat of a security problem, too, since PPPoE sockets can be
created by any user, so any user can easily allocate all the machine's
RAM to non-swappable address space and thus DoS the system.

Is there any specific reason for PPPoE sockets being available to any
unprivileged process, BTW? After all, you need a packet socket for the
discovery stage anyway, so it's unlikely that any unprivileged process
will ever need to create a PPPoE socket, no? Allocating all session IDs
for a known AC is a kind of DoS, too, after all - with Juniper ERXes,
this is really easy, actually, since they don't ever assign session ids
above 8000 ...
Signed-off-by: NFlorian Zumbiehl <florz@florz.de>
Acked-by: NMichal Ostrowski <mostrows@earthlink.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

below you find a patch that (hopefully) fixes a race between an interface
going down and a connect() to a peer on that interface. Before,
connect() would determine that an interface is up, then the interface
could go down and all entries referring to that interface in the
item_hash_table would be marked as ZOMBIEs and their references to
the device would be freed, and after that, connect() would put a new
entry into the hash table referring to the device that meanwhile is
down already - which also would cause unregister_netdevice() to wait
until the socket has been release()d.

This patch does not suffice if we are not allowed to accept connect()s
referring to a device that we already acked a NETDEV_GOING_DOWN for
(that is: all references are only guaranteed to be freed after
NETDEV_DOWN has been acknowledged, not necessarily after the
NETDEV_GOING_DOWN already). And if we are allowed to, we could avoid
looking through the hash table upon NETDEV_GOING_DOWN completely and
only do that once we get the NETDEV_DOWN ...

mostrows:
pppoe_flush_dev is called on NETDEV_GOING_DOWN and NETDEV_DOWN to deal with
this "late connect" issue.  Ideally one would hope to notify users at the
"NETDEV_GOING_DOWN" phase (just to pretend to be nice).  However, it is the
NETDEV_DOWN scan that takes all the responsibility for ensuring nobody is
hanging around at that time.
Signed-off-by: NFlorian Zumbiehl <florz@florz.de>
Acked-by: NMichal Ostrowski <mostrows@earthlink.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

below is a patch that just removes dead code/initializers without any
effect (first access is an assignment) that I stumbled accross while
reading the source.
Signed-off-by: NFlorian Zumbiehl <florz@florz.de>
Acked-by: NMichal Ostrowski <mostrows@earthlink.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Switch cb_lock to mutex and allow netlink kernel users to override it
with a subsystem specific mutex for consistent locking in dump callbacks.
All netlink_dump_start users have been audited not to rely on any
side-effects of the previously used spinlock.
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Rusty added a new 'stats' field to struct net_device.

loopback driver can use it instead of declaring another struct
net_device_stats This saves some memory.
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When csum_offset was introduced we did a conversion from csum to
csum_offset where applicable.  A couple of drivers were missed in
this process.

It was harmless to begin with since the two fields coincided.  Now
that we've made them different with the addition of csum_start, the
missed drivers must be converted or they can't send packets out at
all that require checksum offload.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

To clearly state the intent of copying to linear sk_buffs, _offset being a
overly long variant but interesting for the sake of saving some bytes.
Signed-off-by: NArnaldo Carvalho de Melo <acme@ghostprotocols.net>

We have to put back the cast to "char *" because these
pointers are volatile.

Reported by Andrew Morton.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Network drivers which keep stats allocate their own stats structure
then write a get_stats() function to return them.  It would be nice if
this were done by default.

1) Add a new "stats" field to "struct net_device".
2) Add a new feature field to say "this driver uses the internal one"
3) Have a default "get_stats" which returns NULL if that feature not set.
4) Change callers to check result of get_stats call for NULL, not if
   ->get_stats is set.

This should not break backwards compatibility with older drivers, yet
allow modern drivers to shed some boilerplate code.

Lightly tested: works for a modified lguest network driver.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In order to get more randomness for secure_tcpv6_sequence_number(),
secure_tcp_sequence_number(), secure_dccp_sequence_number() functions,
we can use the high resolution time services, providing nanosec
resolution.

I've also done two kmalloc()/kzalloc() conversions.
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Acked-by: NJames Morris <jmorris@namei.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

To clearly state the intent of copying from linear sk_buffs, _offset being a
overly long variant but interesting for the sake of saving some bytes.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

For consistency with other skb data accessors, reducing the number of direct
accesses to skb->data.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Caused by skb_reset_mac_header() changes, missing semicolon.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Reducing the number of skb->data direct accesses.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

At that point it is equivalent to what was being used, skb->end - skb->data,
and the need is clearly the one skb_tailroom satisfies.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

For the common "(struct nlmsghdr *)skb->data" sequence, so that we reduce the
number of direct accesses to skb->data and for consistency with all the other
cast skb member helpers.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Now to convert the last one, skb->data, that will allow many simplifications
and removal of some of the offset helpers.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

So that it is also an offset from skb->head, reduces its size from 8 to 4 bytes
on 64bit architectures, allowing us to combine the 4 bytes hole left by the
layer headers conversion, reducing struct sk_buff size to 256 bytes, i.e. 4
64byte cachelines, and since the sk_buff slab cache is SLAB_HWCACHE_ALIGN...
:-)

Many calculations that previously required that skb->{transport,network,
mac}_header be first converted to a pointer now can be done directly, being
meaningful as offsets or pointers.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

SMC SuperIO Chip LPC47N227 used for IrDA is not detected because its device
identification byte can be 0x7A instead of 0x5A.

Patch from Peter Kovar <peter.kovar@gmail.com>
Cc: Jean Delvare <khali@linux-fr.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NSamuel Ortiz <samuel@sortiz.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

stir4200 doesn't need to include irlap.h
Signed-off-by: NSamuel Ortiz <samuel@sortiz.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Renaming skb->h to skb->transport_header, skb->nh to skb->network_header and
skb->mac to skb->mac_header, to match the names of the associated helpers
(skb[_[re]set]_{transport,network,mac}_header).
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For the common sequence "skb->h.raw - skb->nh.raw", similar to skb->mac_len,
that is precalculated tho, don't think we need to bloat skb with one more
member, so just use this new helper, reducing the number of non-skbuff.h
references to the layer headers even more.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For the places where we need a pointer to the transport header, it is
still legal to touch skb->h.raw directly if just adding to,
subtracting from or setting it to another layer header.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The ip_hdrlen() buddy, created to reduce the number of skb->h.th-> uses and to
avoid the longer, open coded equivalent.

Ditched a no-op in bnx2 in the process.

I wonder if we should have a BUG_ON(skb->h.th->doff < 5) in tcp_optlen()...
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For the quite common 'skb->h.raw - skb->data' sequence.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For the common, open coded 'skb->h.raw = skb->data' operation, so that we can
later turn skb->h.raw into a offset, reducing the size of struct sk_buff in
64bit land while possibly keeping it as a pointer on 32bit.

This one touches just the most simple cases:

skb->h.raw = skb->data;
skb->h.raw = {skb_push|[__]skb_pull}()

The next ones will handle the slightly more "complex" cases.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Now the skb->nh union has just one member, .raw, i.e. it is just like the
skb->mac union, strange, no? I'm just leaving it like that till the transport
layer is done with, when we'll rename skb->mac.raw to skb->mac_header (or
->mac_header_offset?), ditto for ->{h,nh}.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For the common sequence "skb->nh.iph->ihl * 4", removing a good number of open
coded skb->nh.iph uses, now to go after the rest...

Just out of curiosity, here are the idioms found to get the same result:

skb->nh.iph->ihl << 2
skb->nh.iph->ihl<<2
skb->nh.iph->ihl * 4
skb->nh.iph->ihl*4
(skb->nh.iph)->ihl * sizeof(u32)
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For the places where we need a pointer to the network header, it is still legal
to touch skb->nh.raw directly if just adding to, subtracting from or setting it
to another layer header.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For the quite common 'skb->nh.raw - skb->data' sequence.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For the common, open coded 'skb->nh.raw = skb->data' operation, so that we can
later turn skb->nh.raw into a offset, reducing the size of struct sk_buff in
64bit land while possibly keeping it as a pointer on 32bit.

This one touches just the most simple case, next will handle the slightly more
"complex" cases.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For consistency with all the other skb->nh.raw accessors.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>