openwrt/staging/blogic.git
14 years agonetfilter: xtables: use xt_table for hook instantiation
Jan Engelhardt [Wed, 17 Jun 2009 11:57:48 +0000 (13:57 +0200)]
netfilter: xtables: use xt_table for hook instantiation

The respective xt_table structures already have most of the metadata
needed for hook setup. Add a 'priority' field to struct xt_table so
that xt_hook_link() can be called with a reduced number of arguments.

So should we be having more tables in the future, it comes at no
static cost (only runtime, as before) - space saved:
6807373->6806555.

Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
14 years agonetfilter: xtables: compact table hook functions (2/2)
Jan Engelhardt [Sat, 13 Jun 2009 04:57:10 +0000 (06:57 +0200)]
netfilter: xtables: compact table hook functions (2/2)

The calls to ip6t_do_table only show minimal differences, so it seems
like a good cleanup to merge them to a single one too.
Space saving obtained by both patches: 6807725->6807373
("Total" column from `size -A`.)

Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
14 years agonetfilter: xtables: compact table hook functions (1/2)
Jan Engelhardt [Sat, 13 Jun 2009 04:46:36 +0000 (06:46 +0200)]
netfilter: xtables: compact table hook functions (1/2)

This patch combines all the per-hook functions in a given table into
a single function. Together with the 2nd patch, further
simplifications are possible up to the point of output code reduction.

Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
14 years agoMerge branch 'master' of /repos/git/net-next-2.6
Patrick McHardy [Wed, 10 Feb 2010 13:17:10 +0000 (14:17 +0100)]
Merge branch 'master' of /repos/git/net-next-2.6

Signed-off-by: Patrick McHardy <kaber@trash.net>
14 years agoMerge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
David S. Miller [Tue, 9 Feb 2010 19:44:44 +0000 (11:44 -0800)]
Merge branch 'master' of /linux/kernel/git/davem/net-2.6

14 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/holtmann/bluet...
David S. Miller [Tue, 9 Feb 2010 06:45:56 +0000 (22:45 -0800)]
Merge branch 'master' of git://git./linux/kernel/git/holtmann/bluetooth-2.6

14 years agodrivers/net: Correct NULL test
Julia Lawall [Tue, 9 Feb 2010 06:44:18 +0000 (22:44 -0800)]
drivers/net: Correct NULL test

Test the value that was just allocated rather than the previously tested one.

A simplified version of the semantic match that finds this problem is as
follows: (http://coccinelle.lip6.fr/)

// <smpl>
@r@
expression *x;
expression e;
identifier l;
@@

if (x == NULL || ...) {
    ... when forall
    return ...; }
... when != goto l;
    when != x = e
    when != &x
*x == NULL
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoMAINTAINERS: networking drivers - Add git net-next tree
Joe Perches [Tue, 9 Feb 2010 06:42:40 +0000 (22:42 -0800)]
MAINTAINERS: networking drivers - Add git net-next tree

During the rc period, patches that are not bugfixes
should be done using the net-next tree.

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet/sched: Fix module name in Kconfig
Jan Luebbe [Tue, 9 Feb 2010 06:41:44 +0000 (22:41 -0800)]
net/sched: Fix module name in Kconfig

The action modules have been prefixed with 'act_', but the Kconfig
description was not changed.

Signed-off-by: Jan Luebbe <jluebbe@debian.org>
Acked-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agocxgb3: fix GRO checksum check
Divy Le Ray [Tue, 9 Feb 2010 06:37:24 +0000 (22:37 -0800)]
cxgb3: fix GRO checksum check

Verify the HW checksum state for frames handed to GRO processing.

Signed-off-by: Divy Le Ray <divy@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agodst: call cond_resched() in dst_gc_task()
Eric Dumazet [Mon, 8 Feb 2010 23:00:39 +0000 (15:00 -0800)]
dst: call cond_resched() in dst_gc_task()

Kernel bugzilla #15239

On some workloads, it is quite possible to get a huge dst list to
process in dst_gc_task(), and trigger soft lockup detection.

Fix is to call cond_resched(), as we run in process context.

Reported-by: Pawel Staszewski <pstaszewski@itcare.pl>
Tested-by: Pawel Staszewski <pstaszewski@itcare.pl>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetfilter: nf_conntrack: fix hash resizing with namespaces
Patrick McHardy [Mon, 8 Feb 2010 19:18:07 +0000 (11:18 -0800)]
netfilter: nf_conntrack: fix hash resizing with namespaces

As noticed by Jon Masters <jonathan@jonmasters.org>, the conntrack hash
size is global and not per namespace, but modifiable at runtime through
/sys/module/nf_conntrack/hashsize. Changing the hash size will only
resize the hash in the current namespace however, so other namespaces
will use an invalid hash size. This can cause crashes when enlarging
the hashsize, or false negative lookups when shrinking it.

Move the hash size into the per-namespace data and only use the global
hash size to initialize the per-namespace value when instanciating a
new namespace. Additionally restrict hash resizing to init_net for
now as other namespaces are not handled currently.

Cc: stable@kernel.org
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetfilter: xtables: compat out of scope fix
Alexey Dobriyan [Mon, 8 Feb 2010 19:17:43 +0000 (11:17 -0800)]
netfilter: xtables: compat out of scope fix

As per C99 6.2.4(2) when temporary table data goes out of scope,
the behaviour is undefined:

if (compat) {
struct foo tmp;
...
private = &tmp;
}
[dereference private]

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: stable@kernel.org
Signed-off-by: Patrick McHardy <kaber@trash.net>
14 years agonetfilter: nf_conntrack: restrict runtime expect hashsize modifications
Alexey Dobriyan [Mon, 8 Feb 2010 19:17:22 +0000 (11:17 -0800)]
netfilter: nf_conntrack: restrict runtime expect hashsize modifications

Expectation hashtable size was simply glued to a variable with no code
to rehash expectations, so it was a bug to allow writing to it.
Make "expect_hashsize" readonly.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: stable@kernel.org
Signed-off-by: Patrick McHardy <kaber@trash.net>
14 years agonetfilter: nf_conntrack: per netns nf_conntrack_cachep
Eric Dumazet [Mon, 8 Feb 2010 19:16:56 +0000 (11:16 -0800)]
netfilter: nf_conntrack: per netns nf_conntrack_cachep

nf_conntrack_cachep is currently shared by all netns instances, but
because of SLAB_DESTROY_BY_RCU special semantics, this is wrong.

If we use a shared slab cache, one object can instantly flight between
one hash table (netns ONE) to another one (netns TWO), and concurrent
reader (doing a lookup in netns ONE, 'finding' an object of netns TWO)
can be fooled without notice, because no RCU grace period has to be
observed between object freeing and its reuse.

We dont have this problem with UDP/TCP slab caches because TCP/UDP
hashtables are global to the machine (and each object has a pointer to
its netns).

If we use per netns conntrack hash tables, we also *must* use per netns
conntrack slab caches, to guarantee an object can not escape from one
namespace to another one.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
[Patrick: added unique slab name allocation]
Cc: stable@kernel.org
Signed-off-by: Patrick McHardy <kaber@trash.net>
14 years agonetfilter: nf_conntrack: fix memory corruption with multiple namespaces
Patrick McHardy [Mon, 8 Feb 2010 19:16:26 +0000 (11:16 -0800)]
netfilter: nf_conntrack: fix memory corruption with multiple namespaces

As discovered by Jon Masters <jonathan@jonmasters.org>, the "untracked"
conntrack, which is located in the data section, might be accidentally
freed when a new namespace is instantiated while the untracked conntrack
is attached to a skb because the reference count it re-initialized.

The best fix would be to use a seperate untracked conntrack per
namespace since it includes a namespace pointer. Unfortunately this is
not possible without larger changes since the namespace is not easily
available everywhere we need it. For now move the untracked conntrack
initialization to the init_net setup function to make sure the reference
count is not re-initialized and handle cleanup in the init_net cleanup
function to make sure namespaces can exit properly while the untracked
conntrack is in use in other namespaces.

Cc: stable@kernel.org
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetfilter: fix build failure with CONNTRACK=y NAT=n
Florian Westphal [Mon, 8 Feb 2010 14:39:16 +0000 (15:39 +0100)]
netfilter: fix build failure with CONNTRACK=y NAT=n

net/ipv4/netfilter/nf_defrag_ipv4.c: In function 'ipv4_conntrack_defrag':
net/ipv4/netfilter/nf_defrag_ipv4.c:62: error: implicit declaration of function 'nf_ct_is_template'

Signed-off-by: Florian Westphal <fwestphal@astaro.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
14 years agopacket: Kill CONFIG_PACKET_MMAP.
David S. Miller [Sat, 6 Feb 2010 00:29:48 +0000 (16:29 -0800)]
packet: Kill CONFIG_PACKET_MMAP.

Early on this was an experimental facility that few
people other than Alexey Kuznetsov played with.

Now it's a pretty fundamental thing and as people add
more features to AF_PACKET sockets this config options
creates ifdef spaghetti.

So kill it off.

Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoBluetooth: Keep a copy of each HID device's report descriptor
Michael Poole [Fri, 5 Feb 2010 17:23:43 +0000 (12:23 -0500)]
Bluetooth: Keep a copy of each HID device's report descriptor

The report descriptor is read by user space (via the Service
Discovery Protocol), so it is only available during the ioctl
to connect. However, the HID probe function that needs the
descriptor might not be called until a specific module is
loaded. Keep a copy of the descriptor so it is available for
later use.

Signed-off-by: Michael Poole <mdpoole@troilus.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
14 years agonet: e1000e: convert to use mc helpers
Jiri Pirko [Fri, 5 Feb 2010 02:52:39 +0000 (02:52 +0000)]
net: e1000e: convert to use mc helpers

Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: dm9601: convert to use mc helpers
Jiri Pirko [Fri, 5 Feb 2010 02:51:24 +0000 (02:51 +0000)]
net: dm9601: convert to use mc helpers

Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: 8139too: convert to use mc helpers
Jiri Pirko [Fri, 5 Feb 2010 02:48:11 +0000 (02:48 +0000)]
net: 8139too: convert to use mc helpers

Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: 8139cp: convert to use mc helpers
Jiri Pirko [Fri, 5 Feb 2010 02:47:28 +0000 (02:47 +0000)]
net: 8139cp: convert to use mc helpers

Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: Fix namespace conflicts wrt. e1000_has_link
David S. Miller [Fri, 5 Feb 2010 06:31:41 +0000 (22:31 -0800)]
e1000e: Fix namespace conflicts wrt. e1000_has_link

Reported by Stephen Rothwell.

Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agobridge: Remove unused age_list
Herbert Xu [Fri, 5 Feb 2010 04:28:48 +0000 (20:28 -0800)]
bridge: Remove unused age_list

This patch removes the unused age_list member from the net_bridge
structure.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agopacket: Add GSO/csum offload support.
Sridhar Samudrala [Fri, 5 Feb 2010 04:24:10 +0000 (20:24 -0800)]
packet: Add GSO/csum offload support.

This patch adds GSO/checksum offload to af_packet sockets using
virtio_net_hdr. Based on Rusty's patch to add this support to tun.
It allows GSO/checksum offload to be enabled when using raw socket
backend with virtio_net.
Adds PACKET_VNET_HDR socket option to prepend virtio_net_hdr in the
receive path and process/skip virtio_net_hdr in the send path. This
option is only allowed with SOCK_RAW sockets attached to ethernet
type devices.

v2 updates
----------
Michael's Comments
- Perform length check in packet_snd() when GSO is off even when
  vnet_hdr is present.
- Check for SKB_GSO_FCOE type and return -EINVAL
- don't allow tx/rx ring when vnet_hdr is enabled.
Herbert's Comments
- Removed ethernet specific code.
- protocol value is assumed to be passed in by the caller.

Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlge: Code clean up
Breno Leitao [Thu, 4 Feb 2010 10:11:19 +0000 (10:11 +0000)]
qlge: Code clean up

Just reordering this assignment that doesn't depend on any
condition.

Signed-off-by: Breno Leitao <leitao@linux.vnet.ibm.com>
Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlge: removing unreachable block of code
Breno Leitao [Thu, 4 Feb 2010 10:11:18 +0000 (10:11 +0000)]
qlge: removing unreachable block of code

Currently the qlge_change_mtu() is never called if the new_mtu is
equal current MTU, due this condition on dev_set_mtu():

        if (new_mtu == dev->mtu)
                return 0;

So, this block of code is never reached and is being removed.

Signed-off-by: Breno Leitao <leitao@linux.vnet.ibm.com>
Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoipv4: obsolete config in kernel source (IP_ROUTE_PERVASIVE)
Christoph Egger [Wed, 3 Feb 2010 06:05:54 +0000 (06:05 +0000)]
ipv4: obsolete config in kernel source (IP_ROUTE_PERVASIVE)

CONFIG_IP_ROUTE_PERVASIVE is missing a corresponding config
IP_ROUTE_PERVASIVE somewhere in KConfig (and missing it for ages
already) so it looks like some aging artefact no longer needed.

Therefor this patch kills of the only remaining reference to that
config Item removing the already unrechable code snipet.

Signed-off-by: Christoph Egger <siccegge@stud.informatik.uni-erlangen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agopktgen: Fix freezing problem
Rafael J. Wysocki [Thu, 4 Feb 2010 22:00:41 +0000 (14:00 -0800)]
pktgen: Fix freezing problem

Add missing try_to_freeze() to one of the pktgen_thread_worker() code
paths so that it doesn't block suspend/hibernation.

Fixes http://bugzilla.kernel.org/show_bug.cgi?id=15006

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Reported-and-tested-by: Ciprian Dorin Craciun <ciprian.craciun@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agodoc: document IPv6 parameters
Brian Haley [Thu, 4 Feb 2010 21:36:50 +0000 (13:36 -0800)]
doc: document IPv6 parameters

Update documentation to describe IPv6 parameters.
Reported by <greg@enjellic.com>.

Signed-off-by: Brian Haley <brian.haley@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlge: Add watchdog timer.
Ron Mercer [Thu, 4 Feb 2010 21:32:46 +0000 (13:32 -0800)]
qlge: Add watchdog timer.

Add periodic heartbeat register read to trigger the eeh
recovery process.
We see cases where an eeh error was injected and the slot was
suspended.  An asic access attempt is required to flush the recovery process,
but without interrupts the process can stall.
Adding this periodic register read causes the recovery process to begin.

Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agolibphy: add phy_find_first function
Jiri Pirko [Thu, 4 Feb 2010 18:23:02 +0000 (10:23 -0800)]
libphy: add phy_find_first function

Many drivers do this in them manually. Now they can use this function.

Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: use helpers to access mc list V2
Jiri Pirko [Thu, 4 Feb 2010 18:22:25 +0000 (10:22 -0800)]
net: use helpers to access mc list V2

This patch introduces the similar helpers as those already done for uc list.
However multicast lists are no list_head lists but "mademanually". The three
macros added by this patch will make the transition of mc_list to list_head
smooth in two steps:

1) convert all drivers to use these macros (with the original iterator of type
   "struct dev_mc_list")
2) once all drivers are converted, convert list type and iterators to "struct
   netdev_hw_addr" in one patch.

>From now on, drivers can (and should) use "netdev_for_each_mc_addr" to iterate
over the addresses with iterator of type "struct netdev_hw_addr". Also macros
"netdev_mc_count" and "netdev_mc_empty" to read list's length. This is the state
which should be reached in all drivers.

Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoigb: make certain to reassign legacy interrupt vectors after reset
Alexander Duyck [Wed, 3 Feb 2010 21:59:51 +0000 (21:59 +0000)]
igb: make certain to reassign legacy interrupt vectors after reset

This change corrects an issue that will cause false hangs when using either
82575 or 82580 in legacy interrupt mode.  The issue is caused when there is
a slow traffic flow and an "ethtool -r" is executed while using legacy or
MSI interrupts.  MSI-X is not affected by this issue due to the fact that
we were already reconfiguring the vectors after reset.

If possible it would be best to push this for net-2.6 since it is resolving
a bug but if that is not possible then net-next-2.6 will be fine.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agostmmac: fix 'lenght' typo in comments and code
Giuseppe Cavallaro [Thu, 4 Feb 2010 17:33:21 +0000 (09:33 -0800)]
stmmac: fix 'lenght' typo in comments and code

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agocan: ems_usb: removed duplicated code setting local echo support
Thadeu Lima de Souza Cascardo [Thu, 4 Feb 2010 17:21:28 +0000 (09:21 -0800)]
can: ems_usb: removed duplicated code setting local echo support

Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wirel...
David S. Miller [Thu, 4 Feb 2010 16:58:14 +0000 (08:58 -0800)]
Merge branch 'master' of git://git./linux/kernel/git/linville/wireless-next-2.6

14 years agosky2: Flow control frames recorded as dropped packets
Stephen Hemminger [Wed, 3 Feb 2010 08:31:12 +0000 (08:31 +0000)]
sky2: Flow control frames recorded as dropped packets

Thanks for your patch.  A more general solution would be to move the
rx_dropped up into sky2_receive.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosky2: hand receive DMA mapping failures
stephen hemminger [Mon, 1 Feb 2010 13:45:41 +0000 (13:45 +0000)]
sky2: hand receive DMA mapping failures

If receive buffer mapping failed, then it was possible to get
stuck with unmapped receive buffer in DMA ring.

This would be an extremely rare condition because the driver had just
released the map for the last receive so it should be able to get
another map again (in soft-irq).

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: CONFIG_COMPAT redux
Alexey Dobriyan [Mon, 1 Feb 2010 09:44:19 +0000 (09:44 +0000)]
net: CONFIG_COMPAT redux

Ifdef out
struct proto_ops::compat_ioctl
struct proto_ops::compat_setsockopt
struct proto_ops::compat_getsockopt
to make structures smaller on COMPAT=n kernels.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agocan: mcp251x: Move to threaded interrupts instead of workqueues.
Christian Pellegrin [Wed, 3 Feb 2010 07:39:54 +0000 (07:39 +0000)]
can: mcp251x: Move to threaded interrupts instead of workqueues.

This patch addresses concerns about efficiency of handling incoming
packets. Handling of interrupts is done in a threaded interrupt handler
which has a smaller latency than workqueues. This change needed a rework
of the locking scheme that was much simplified. Some other (more or less
longstanding) bugs are fixed: utilization of just half of the RX
buffers, useless wait for interrupt on open, more reliable reset
sequence. The MERR interrupt is not used anymore: it overloads the CPU
in error-passive state without any additional information. One shot mode
is disabled because it's not clear if it can be handled efficiently on
this CAN controller.

Signed-off-by: Christian Pellegrin <chripell@fsfe.org>
Acked-by: Wolfgang Grandegger <wg@grandegger.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agobonding: Remove net_device_stats from bonding struct
Ajit Khaparde [Mon, 1 Feb 2010 14:06:52 +0000 (14:06 +0000)]
bonding: Remove net_device_stats from bonding struct

There is no need to maintain stats in the bonding structure.
Use the instance of net_device_stats in netdevice.

Signed-off-by: Ajit Khaparde <ajitk@serverengines.com>
Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agodrivers/net/davinci_emac.c: Fix continuation line formats
Joe Perches [Mon, 1 Feb 2010 21:22:11 +0000 (21:22 +0000)]
drivers/net/davinci_emac.c: Fix continuation line formats

String constants that are continued on subsequent lines with \
are not good.

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agotrivial: remove duplicated "from" in CAN USB EMS Kconfig help
Thadeu Lima de Souza Cascardo [Mon, 1 Feb 2010 07:58:59 +0000 (07:58 +0000)]
trivial: remove duplicated "from" in CAN USB EMS Kconfig help

Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: macvtap driver
Arnd Bergmann [Sat, 30 Jan 2010 12:24:26 +0000 (12:24 +0000)]
net: macvtap driver

In order to use macvlan with qemu and other tools that require
a tap file descriptor, the macvtap driver adds a small backend
with a character device with the same interface as the tun
driver, with a minimum set of features.

Macvtap interfaces are created in the same way as macvlan
interfaces using ip link, but the netif is just used as a
handle for configuration and accounting, while the data
goes through the chardev. Each macvtap interface has its
own character device, simplifying permission management
significantly over the generic tun/tap driver.

Cc: Patrick McHardy <kaber@trash.net>
Cc: Stephen Hemminger <shemminger@linux-foundation.org>
Cc: David S. Miller" <davem@davemloft.net>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Or Gerlitz <ogerlitz@voltaire.com>
Cc: netdev@vger.kernel.org
Cc: bridge@lists.linux-foundation.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agomacvlan: allow multiple driver backends
Arnd Bergmann [Sat, 30 Jan 2010 12:23:40 +0000 (12:23 +0000)]
macvlan: allow multiple driver backends

This makes it possible to hook into the macvlan driver
from another kernel module. In particular, the goal is
to extend it with the macvtap backend that provides
a tun/tap compatible interface directly on the macvlan
device.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: maintain namespace isolation between vlan and real device
Arnd Bergmann [Sat, 30 Jan 2010 12:23:03 +0000 (12:23 +0000)]
net: maintain namespace isolation between vlan and real device

In the vlan and macvlan drivers, the start_xmit function forwards
data to the dev_queue_xmit function for another device, which may
potentially belong to a different namespace.

To make sure that classification stays within a single namespace,
this resets the potentially critical fields.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet/rds: remove uses of NIPQUAD, use %pI4
Joe Perches [Tue, 2 Feb 2010 12:43:59 +0000 (12:43 +0000)]
net/rds: remove uses of NIPQUAD, use %pI4

Signed-off-by: Joe Perches <joe@perches.com>
Cc: Andy Grover <andy.grover@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agomISDN: positive error return should be negative in mode_hfcmulti()
Roel Kluin [Tue, 2 Feb 2010 12:43:47 +0000 (12:43 +0000)]
mISDN: positive error return should be negative in mode_hfcmulti()

The error return should be negative.  Its only caller that acts upon its
return, handle_bmsg(), transmits the positive error but can also return
negative errors.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Cc: Karsten Keil <isdn@linux-pingi.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoisdn: mISDN, don't compile unused stuff
Jiri Slaby [Tue, 2 Feb 2010 12:43:46 +0000 (12:43 +0000)]
isdn: mISDN, don't compile unused stuff

Remove these compiler warnings:

drivers/isdn/hardware/mISDN/w6692.c:534: warning: `setvolume' defined but not used
drivers/isdn/hardware/mISDN/w6692.c:561: warning: `enable_pots' defined but not used

by moving the functions inside #if 0 ...  #endif.  And an alternative is
to remove them completely if nobody has plans to use them.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: Karsten Keil <isdn@linux-pingi.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agohardware/mISDN/mISDNinfineon.c: bail out of loop on error
Darren Jenkins [Tue, 2 Feb 2010 12:43:45 +0000 (12:43 +0000)]
hardware/mISDN/mISDNinfineon.c: bail out of loop on error

If setup_instance() fails we kfree() the card, and then use it in the next
loop iteration.  So lets bail out of the loop instead.

Coverity CID: 13357

Signed-off-by: Darren Jenkins <darrenrjenkins@gmail.com>
Cc: Karsten Keil <isdn@linux-pingi.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agohisax: timeout off by one in waitrecmsg()
Roel Kluin [Tue, 2 Feb 2010 12:43:44 +0000 (12:43 +0000)]
hisax: timeout off by one in waitrecmsg()

With `while (timeout++ < maxdelay)' timeout reaches maxdelay + 1 after the
loop This is probably unlikely a problem in practice.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Cc: Karsten Keil <isdn@linux-pingi.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoirda: add missing BKL in irnet_ppp ioctl
Thadeu Lima de Souza Cascardo [Mon, 1 Feb 2010 08:21:35 +0000 (08:21 +0000)]
irda: add missing BKL in irnet_ppp ioctl

One ioctl has been forgotten when the BKL was push down into irnet_ppp
ioctl function.

Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoirda: unbalanced lock_kernel in irnet_ppp
Thadeu Lima de Souza Cascardo [Mon, 1 Feb 2010 08:21:34 +0000 (08:21 +0000)]
irda: unbalanced lock_kernel in irnet_ppp

Add the missing unlock_kernel in one ioctl operation.

Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlge: Add check for eeh failure when closing device.
Ron Mercer [Wed, 3 Feb 2010 07:24:12 +0000 (07:24 +0000)]
qlge: Add check for eeh failure when closing device.

Fix crash where resources are freed twice on an eeh recovery failure.
If eeh recovery fails we set a flag to indicate to close() that
resources have been freed.

Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlge: Move reset from eeh io_resume to slot_reset.
Ron Mercer [Wed, 3 Feb 2010 07:24:11 +0000 (07:24 +0000)]
qlge: Move reset from eeh io_resume to slot_reset.

Issue asic reset and verify functionality before continuing to the
resume call.  This allows proper error code to be returned in the case
the asic does not recover.

Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoixgbe: Make descriptor ring allocations NUMA-aware
PJ Waskiewicz [Wed, 3 Feb 2010 14:19:12 +0000 (14:19 +0000)]
ixgbe: Make descriptor ring allocations NUMA-aware

This patch allocates the ring structures themselves on each
NUMA node along with the buffer_info structures.  This way we
don't allocate the entire ring memory on a single node in one
big block, thus reducing NUMA node memory crosstalk.

Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoixgbe: Allocate driver resources per NUMA node
Jesse Brandeburg [Wed, 3 Feb 2010 14:18:50 +0000 (14:18 +0000)]
ixgbe: Allocate driver resources per NUMA node

The default policy for the current driver is to do all its memory
allocation on whatever processor is running insmod/modprobe.  This
is less than optimal.

This driver's default mode of operation will be to use each node for each
subsequent transmit/receive queue.  The most efficient allocation will be
to then have the interrupts bound in such a way as to match up the
interrupt of the queue to the cpu where its memory was allocated.

Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000: call pci_save_state after pci_restore_state
Nick Nunley [Wed, 3 Feb 2010 14:49:48 +0000 (14:49 +0000)]
e1000: call pci_save_state after pci_restore_state

This patch adds a call to pci_save_state() immediately after
the call to pci_restore_state(). Due to a change in the behavior
of pci_restore_state() it is necessary to call pci_save_state()
to keep the state_saved flag. This patch is based on a similar
patch for ixgbe.

Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000: Report link status in ethtool when interface is down
Nick Nunley [Wed, 3 Feb 2010 14:49:28 +0000 (14:49 +0000)]
e1000: Report link status in ethtool when interface is down

With this change ethtool will correctly report link status when
the interface is down. Currently ethtool reports the link as not
detected when the interface is down.

Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoixgbe: only process one ixgbe_watchdog_task at a time.
John Fastabend [Wed, 3 Feb 2010 14:23:32 +0000 (14:23 +0000)]
ixgbe: only process one ixgbe_watchdog_task at a time.

Processing multiple ixgbe_watchdog_task calls may cause
the link_up variable and IXGBE_FLAG_NEED_LINK_UPDATE flag
to be set incorrectly.  In the worse case this is causing
the netif_carrier_off to be called inappropriately which
results in an interface that can't be brought up.

Although schedule_work() will only schedule the task if
it is not already on the work queue the WORK_STRUCT_PENDING
bits are cleared just before calling the work function.
This allows WORK_STRUCT_PENDING to be cleared, the work
function to start and meanwhile schedule another task.

This patch adds a mutex to the watchdog task. This bug is
actualized by changing DCB settings or doing extended
cable pull or reset tests.

Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoMerge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
David S. Miller [Thu, 4 Feb 2010 03:38:22 +0000 (19:38 -0800)]
Merge branch 'master' of /linux/kernel/git/davem/net-2.6

14 years agoixgbe: Fix return of invalid txq
Krishna Kumar [Wed, 3 Feb 2010 13:13:10 +0000 (13:13 +0000)]
ixgbe: Fix return of invalid txq

a developer had complained of getting lots of warnings:

"eth16 selects TX queue 98, but real number of TX queues is 64"

http://www.mail-archive.com/e1000-devel@lists.sourceforge.net/msg02200.html

As there was no follow up on that bug, I am submitting this
patch assuming that the other return points will not return
invalid txq's, and also that this fixes the bug (not tested).

Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Acked-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoixgbe: Fix ixgbe_tx_map error path
Anton Blanchard [Wed, 3 Feb 2010 13:12:51 +0000 (13:12 +0000)]
ixgbe: Fix ixgbe_tx_map error path

Commit e5a43549f7a58509a91b299a51337d386697b92c (ixgbe: remove
skb_dma_map/unmap calls from driver) looks to have introduced a bug in
ixgbe_tx_map. If we get an error from a PCI DMA call, we loop backwards
through count until it becomes -1 and return that.

The caller of ixgbe_tx_map expects 0 on error, so return that instead.

Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetxen: protect resource cleanup by rtnl lock
Amit Kumar Salecha [Tue, 2 Feb 2010 04:16:21 +0000 (04:16 +0000)]
netxen: protect resource cleanup by rtnl lock

o context resources can be in used, while resource cleanup is in progress,
  during fw recover.
o Null pointer execption can occur in send_cmd_desc, if fw recovery
  module frees tx ring without rtnl lock.
o Same applies to ethtool register dump.

Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetxen: fix tx timeout recovery for NX2031 chip
Amit Kumar Salecha [Tue, 2 Feb 2010 04:16:20 +0000 (04:16 +0000)]
netxen: fix tx timeout recovery for NX2031 chip

For NX2031, first try to scrub interrupt before requesting firmware
reset. Return statement was missing after scrubbbing interrupt.

Signed-off-by: Vernon Mauery <vernux@us.ibm.com>
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosfc: Do not include unneeded headers
Ben Hutchings [Wed, 3 Feb 2010 09:32:06 +0000 (09:32 +0000)]
sfc: Do not include unneeded headers

Earlier refactoring has made these inclusions unnecessary.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosfc: Fix some incorrect or redundant comments
Ben Hutchings [Wed, 3 Feb 2010 09:31:57 +0000 (09:31 +0000)]
sfc: Fix some incorrect or redundant comments

In particular, the comment about EVQ_RPTR_REG is based on inconsistent
preliminary hardware documentation, though the following code was
fixed long before release.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosfc: Remove declarations of nonexistent functions
Ben Hutchings [Wed, 3 Feb 2010 09:31:46 +0000 (09:31 +0000)]
sfc: Remove declarations of nonexistent functions

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosfc: Add some missing bits to register self-test masks
Steve Hodgson [Wed, 3 Feb 2010 09:31:40 +0000 (09:31 +0000)]
sfc: Add some missing bits to register self-test masks

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosfc: Survive ISR0=0 bug in the shared IRQ case
Guido Barzini [Wed, 3 Feb 2010 09:31:24 +0000 (09:31 +0000)]
sfc: Survive ISR0=0 bug in the shared IRQ case

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosfc: Implement NVRAM selftest for SFC9000 family
Ben Hutchings [Wed, 3 Feb 2010 09:31:01 +0000 (09:31 +0000)]
sfc: Implement NVRAM selftest for SFC9000 family

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosfc: Replace PHY MDIO test with an 'alive' test
Ben Hutchings [Wed, 3 Feb 2010 09:30:50 +0000 (09:30 +0000)]
sfc: Replace PHY MDIO test with an 'alive' test

SFC9000-family boards do not all use MDIO PHYs, so we need a different
test for PHY aliveness.

Introduce a PHY operation test_alive().  For PHYs attached to Falcon,
use a common implementation based on the existing PHY MDIO test.
For PHYs managed through MCDI, use the appropriate MCDI request.

Change test name in ethtool from 'core mdio' to 'phy alive'.

Rename test_results::mdio to phy_alive and test_results::phy to phy_ext.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosfc: Enable autonegotiated flow-control by default if supported
Steve Hodgson [Wed, 3 Feb 2010 09:30:38 +0000 (09:30 +0000)]
sfc: Enable autonegotiated flow-control by default if supported

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosfc: Handle firmware assertion failure while resetting
Steve Hodgson [Wed, 3 Feb 2010 09:30:17 +0000 (09:30 +0000)]
sfc: Handle firmware assertion failure while resetting

This allows the driver to recover if the MC firmware has crashed due
to an assertion failure.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosfc: Update MCDI protocol definitions
Ben Hutchings [Wed, 3 Feb 2010 09:28:14 +0000 (09:28 +0000)]
sfc: Update MCDI protocol definitions

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoBluetooth: Enter active mode before establishing a SCO link.
Nick Pelly [Fri, 13 Nov 2009 22:16:32 +0000 (14:16 -0800)]
Bluetooth: Enter active mode before establishing a SCO link.

When in sniff mode with a long interval time (1.28s) it can take 4+ seconds
to establish a SCO link. Fix by requesting active mode before requesting
SCO connection. This improves SCO setup time to ~500ms.

Bluetooth headsets that use a long interval time, and exhibit the long
SCO connection time include Motorola H790, HX1 and H17. They have a
CSR 2.1 chipset.

Verified this behavior and fix with host Bluetooth chipsets: BCM4329 and
TI1271.

2009-10-13 14:17:46.183722 > HCI Event: Mode Change (0x14) plen 6
    status 0x00 handle 1 mode 0x02 interval 2048
    Mode: Sniff
2009-10-13 14:17:53.436285 < HCI Command: Setup Synchronous Connection (0x01|0x0028) plen 17
    handle 1 voice setting 0x0060
2009-10-13 14:17:53.445593 > HCI Event: Command Status (0x0f) plen 4
    Setup Synchronous Connection (0x01|0x0028) status 0x00 ncmd 1
2009-10-13 14:17:57.788855 > HCI Event: Synchronous Connect Complete 0x2c) plen 17
    status 0x00 handle 257 bdaddr 00:1A:0E:F1:A4:7F type eSCO
    Air mode: CVSD

Signed-off-by: Nick Pelly <npelly@google.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
14 years agoBluetooth: Fix memory leak in Marvell BT-over-SDIO driver
Yoichi Yuasa [Thu, 4 Feb 2010 00:55:51 +0000 (09:55 +0900)]
Bluetooth: Fix memory leak in Marvell BT-over-SDIO driver

Signed-off-by: Yoichi Yuasa <yuasa@linux-mips.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
14 years agodccp: fix auto-loading of dccp(_probe)
Gerrit Renker [Tue, 2 Feb 2010 20:16:56 +0000 (20:16 +0000)]
dccp: fix auto-loading of dccp(_probe)

This fixes commit (38ff3e6bb987ec583268da8eb22628293095d43b) ("dccp_probe:
Fix module load dependencies between dccp and dccp_probe", from 15 Jan).

It fixes the construction of the first argument of try_then_request_module(),
where only valid return codes from the first argument should be returned.

What we do now is assign the result of register_jprobe() to ret, without
the side effect of the comparison.

Acked-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agodccp: fix bug in cache allocation
Gerrit Renker [Mon, 1 Feb 2010 02:12:19 +0000 (02:12 +0000)]
dccp: fix bug in cache allocation

This fixes a bug introduced in commit de4ef86cfce60d2250111f34f8a084e769f23b16
("dccp: fix dccp rmmod when kernel configured to use slub", 17 Jan): the
vsnprintf used sizeof(slab_name_fmt), which became truncated to 4 bytes, since
slab_name_fmt is now a 4-byte pointer and no longer a 32-character array.

This lead to error messages such as
 FATAL: Error inserting dccp: No buffer space available

 >> kernel: [ 1456.341501] kmem_cache_create: duplicate cache cci
generated due to the truncation after the 3rd character.

Fixed for the moment by introducing a symbolic constant. Tested to fix the bug.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agodrivers/net/amd8111e.c: Fix continuation line formats
Joe Perches [Sun, 31 Jan 2010 10:02:09 +0000 (10:02 +0000)]
drivers/net/amd8111e.c: Fix continuation line formats

String constants that are continued on subsequent lines with \
are not good.

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agocan: add support for CAN interface cards based on the PLX90xx PCI bridge
Pavel Cheblakov [Mon, 1 Feb 2010 09:42:44 +0000 (09:42 +0000)]
can: add support for CAN interface cards based on the PLX90xx PCI bridge

This driver is for CAN interface cards based on the PLX90xx PCI bridge.
Driver supports now:
 - Adlink PCI-7841/cPCI-7841 card (http://www.adlinktech.com/)
 - Adlink PCI-7841/cPCI-7841 SE card
 - Marathon CAN-bus-PCI card (http://www.marathon.ru/)
 - TEWS TECHNOLOGIES TPMC810 card (http://www.tews.com/)

Changes since v1:
 - Added some defines for static inline int plx_pci_check_sja1000(...)
 - static struct pci_device_id plx_pci_tbl[] replaced by
   static DEFINE_PCI_DEVICE_TABLE(plx_pci_tbl)
 - Typo fixed

Signed-off-by: Pavel Cheblakov <P.B.Cheblakov@inp.nsk.su>
Acked-by: Wolfgang Grandegger <wg@grandegger.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agocxgb3: add memory barriers
Divy Le Ray [Mon, 1 Feb 2010 10:29:29 +0000 (10:29 +0000)]
cxgb3: add memory barriers

Add memory barriers to fix crashes observed on newest PowerPC platforms.
The HW and driver state of the receive rings were getting out of sync.

Signed-off-by: Divy Le Ray <divy@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosky2: fix transmit DMA map leakage
stephen hemminger [Mon, 1 Feb 2010 13:41:47 +0000 (13:41 +0000)]
sky2: fix transmit DMA map leakage

The book keeping structure for transmit always had the flags value
cleared so transmit DMA maps were never released correctly.
Based on patch by Jarek Poplawski, problem observed by Michael Breuer.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoMCS7830 USB-Ether: resume _with_ working link, via .reset_resume support
Andreas Mohr [Sun, 31 Jan 2010 02:58:42 +0000 (02:58 +0000)]
MCS7830 USB-Ether: resume _with_ working link, via .reset_resume support

ChangeLog:
Implement .reset_resume support to retain a live network connection
during suspend despite USB power loss.
- rework operation to reference cached data in mcs7830_data and
  netdev->dev_addr
- update netdev->dev_addr only in case new MAC was set successfully
. Tests done:
  . ethtool -d pre-/post-suspend: register values match
  . running ssh session suspend, resume: works
  . ifdown device, suspend, resume: works
  . ifup, suspend, unplug, resume: WORKS (eth1 is removed, re-ifup of eth1
    after card replug works)
  . verified identical MAC in ifconfig post-resume
    (ok, should be verified on network side to be fully certain...)

Keywords: suspend resume network connection dead interface down

Signed-off-by: Andreas Mohr <andi@lisas.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoMCS7830 USB-Ether: change register define
Andreas Mohr [Sun, 31 Jan 2010 02:58:33 +0000 (02:58 +0000)]
MCS7830 USB-Ether: change register define

ChangeLog:
- rename register, add comment

Signed-off-by: Andreas Mohr <andi@lisas.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoMCS7830 USB-Ether: Spelling corrections
Andreas Mohr [Sun, 31 Jan 2010 02:58:26 +0000 (02:58 +0000)]
MCS7830 USB-Ether: Spelling corrections

ChangeLog:
- spelling corrections / whitespace

Signed-off-by: Andreas Mohr <andi@lisas.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoMCS7830 USB-Ether: add Rx error support
Andreas Mohr [Sun, 31 Jan 2010 02:58:19 +0000 (02:58 +0000)]
MCS7830 USB-Ether: add Rx error support

ChangeLog:
- evaluate Rx error statistics from trailing Rx status byte
- add driver TODO list
- add myself to authors

Quilt series run-tested, based on 2.6.33-rc4 (net-2.6.git mcs7830 has idle history,
should be good to go).

Signed-off-by: Andreas Mohr <andi@lisas.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetlink: fix for too early rmmod
Alexey Dobriyan [Sat, 30 Jan 2010 10:05:05 +0000 (10:05 +0000)]
netlink: fix for too early rmmod

Netlink code does module autoload if protocol userspace is asking for is
not ready. However, module can dissapear right after it was autoloaded.
Example: modprobe/rmmod stress-testing and xfrm_user.ko providing NETLINK_XFRM.

netlink_create() in such situation _will_ create userspace socket and
_will_not_ pin module. Now if module was removed and we're going to call
->netlink_rcv into nothing:

BUG: unable to handle kernel paging request at ffffffffa02f842a
       ^^^^^^^^^^^^^^^^
modules are loaded near these addresses here

IP: [<ffffffffa02f842a>] 0xffffffffa02f842a
PGD 161f067 PUD 1623063 PMD baa12067 PTE 0
Oops: 0010 [#1] PREEMPT SMP DEBUG_PAGEALLOC
last sysfs file: /sys/devices/pci0000:00/0000:00:1f.2/host0/target0:0:0/0:0:0:0/block/sda/uevent
CPU 1
Pid: 11515, comm: ip Not tainted 2.6.33-rc5-netns-00594-gaaa5728-dirty #6 P5E/P5E
RIP: 0010:[<ffffffffa02f842a>]  [<ffffffffa02f842a>] 0xffffffffa02f842a
RSP: 0018:ffff8800baa3db48  EFLAGS: 00010292
RAX: ffff8800baa3dfd8 RBX: ffff8800be353640 RCX: 0000000000000000
RDX: ffffffff81959380 RSI: ffff8800bab7f130 RDI: 0000000000000001
RBP: ffff8800baa3db58 R08: 0000000000000001 R09: 0000000000000000
R10: 0000000000000001 R11: 0000000000000001 R12: 0000000000000011
R13: ffff8800be353640 R14: ffff8800bcdec240 R15: ffff8800bd488010
FS:  00007f93749656f0(0000) GS:ffff880002300000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: ffffffffa02f842a CR3: 00000000ba82b000 CR4: 00000000000006e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process ip (pid: 11515, threadinfo ffff8800baa3c000, task ffff8800bab7eb30)
Stack:
 ffffffff813637c0 ffff8800bd488000 ffff8800baa3dba8 ffffffff8136397d
<0> 0000000000000000 ffffffff81344adc 7fffffffffffffff 0000000000000000
<0> ffff8800baa3ded8 ffff8800be353640 ffff8800bcdec240 0000000000000000
Call Trace:
 [<ffffffff813637c0>] ? netlink_unicast+0x100/0x2d0
 [<ffffffff8136397d>] netlink_unicast+0x2bd/0x2d0

netlink_unicast_kernel:
nlk->netlink_rcv(skb);

 [<ffffffff81344adc>] ? memcpy_fromiovec+0x6c/0x90
 [<ffffffff81364263>] netlink_sendmsg+0x1d3/0x2d0
 [<ffffffff8133975b>] sock_sendmsg+0xbb/0xf0
 [<ffffffff8106cdeb>] ? __lock_acquire+0x27b/0xa60
 [<ffffffff810a18c3>] ? might_fault+0x73/0xd0
 [<ffffffff810a18c3>] ? might_fault+0x73/0xd0
 [<ffffffff8106db22>] ? __lock_release+0x82/0x170
 [<ffffffff810a190e>] ? might_fault+0xbe/0xd0
 [<ffffffff810a18c3>] ? might_fault+0x73/0xd0
 [<ffffffff81344c77>] ? verify_iovec+0x47/0xd0
 [<ffffffff8133a509>] sys_sendmsg+0x1a9/0x360
 [<ffffffff813c2be5>] ? _raw_spin_unlock_irqrestore+0x65/0x70
 [<ffffffff8106aced>] ? trace_hardirqs_on+0xd/0x10
 [<ffffffff813c2bc2>] ? _raw_spin_unlock_irqrestore+0x42/0x70
 [<ffffffff81197004>] ? __up_read+0x84/0xb0
 [<ffffffff8106ac95>] ? trace_hardirqs_on_caller+0x145/0x190
 [<ffffffff813c207f>] ? trace_hardirqs_on_thunk+0x3a/0x3f
 [<ffffffff8100262b>] system_call_fastpath+0x16/0x1b
Code:  Bad RIP value.
RIP  [<ffffffffa02f842a>] 0xffffffffa02f842a
 RSP <ffff8800baa3db48>
CR2: ffffffffa02f842a

If module was quickly removed after autoloading, return -E.

Return -EPROTONOSUPPORT if module was quickly removed after autoloading.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoaf_key: fix netns ops ordering on module load/unload
Alexey Dobriyan [Sat, 30 Jan 2010 02:53:27 +0000 (02:53 +0000)]
af_key: fix netns ops ordering on module load/unload

1. After sock_register() returns, it's possible to create sockets,
   even if module still not initialized fully (blame generic module code
   for that!)
2. Consequently, pfkey_create() can be called with pfkey_net_id still not
   initialized which will BUG_ON in net_generic():
kernel BUG at include/net/netns/generic.h:43!
3. During netns shutdown, netns ops should be unregistered after
   key manager unregistered because key manager calls can be triggered
   from xfrm_user module:

    general protection fault: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC
pfkey_broadcast+0x111/0x210 [af_key]
pfkey_send_notify+0x16a/0x300 [af_key]
km_state_notify+0x41/0x70
xfrm_flush_sa+0x75/0x90 [xfrm_user]
4. Unregister netns ops after socket ops just in case and for symmetry.

Reported by Luca Tettamanti.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Tested-by: Luca Tettamanti <kronos.it@gmail.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoBluetooth: Do not call rfcomm_session_put() for RFCOMM UA on closed socket
Nick Pelly [Thu, 4 Feb 2010 00:18:36 +0000 (16:18 -0800)]
Bluetooth: Do not call rfcomm_session_put() for RFCOMM UA on closed socket

When processing a RFCOMM UA frame when the socket is closed and we were
not the RFCOMM initiator would cause rfcomm_session_put() to be called
twice during rfcomm_process_rx(). This would cause a kernel panic in
rfcomm_session_close() then.

This could be easily reproduced during disconnect with devices such as
Motorola H270 that send RFCOMM UA followed quickly by L2CAP disconnect
request. This trace for this looks like:

2009-09-21 17:22:37.788895 < ACL data: handle 1 flags 0x02 dlen 8
   L2CAP(d): cid 0x0041 len 4 [psm 3]
     RFCOMM(s): DISC: cr 0 dlci 20 pf 1 ilen 0 fcs 0x7d
2009-09-21 17:22:37.906204 > HCI Event: Number of Completed Packets (0x13) plen 5
   handle 1 packets 1
2009-09-21 17:22:37.933090 > ACL data: handle 1 flags 0x02 dlen 8
   L2CAP(d): cid 0x0040 len 4 [psm 3]
     RFCOMM(s): UA: cr 0 dlci 20 pf 1 ilen 0 fcs 0x57
2009-09-21 17:22:38.636764 < ACL data: handle 1 flags 0x02 dlen 8
   L2CAP(d): cid 0x0041 len 4 [psm 3]
     RFCOMM(s): DISC: cr 0 dlci 0 pf 1 ilen 0 fcs 0x9c
2009-09-21 17:22:38.744125 > HCI Event: Number of Completed Packets (0x13) plen 5
   handle 1 packets 1
2009-09-21 17:22:38.763687 > ACL data: handle 1 flags 0x02 dlen 8
   L2CAP(d): cid 0x0040 len 4 [psm 3]
     RFCOMM(s): UA: cr 0 dlci 0 pf 1 ilen 0 fcs 0xb6
2009-09-21 17:22:38.783554 > ACL data: handle 1 flags 0x02 dlen 12
   L2CAP(s): Disconn req: dcid 0x0040 scid 0x0041

Avoid calling rfcomm_session_put() twice by skipping this call
in rfcomm_recv_ua() if the socket is closed.

Signed-off-by: Nick Pelly <npelly@google.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
14 years agoBluetooth: Fix sleeping function in RFCOMM within invalid context
Marcel Holtmann [Wed, 3 Feb 2010 23:52:18 +0000 (15:52 -0800)]
Bluetooth: Fix sleeping function in RFCOMM within invalid context

With the commit 9e726b17422bade75fba94e625cd35fd1353e682 the
rfcomm_session_put() gets accidentially called from a timeout
callback and results in this:

BUG: sleeping function called from invalid context at net/core/sock.c:1897
in_atomic(): 1, irqs_disabled(): 0, pid: 0, name: swapper
Pid: 0, comm: swapper Tainted: P           2.6.32 #31
Call Trace:
 <IRQ>  [<ffffffff81036455>] __might_sleep+0xf8/0xfa
 [<ffffffff8138ef1d>] lock_sock_nested+0x29/0xc4
 [<ffffffffa03921b3>] lock_sock+0xb/0xd [l2cap]
 [<ffffffffa03948e6>] l2cap_sock_shutdown+0x1c/0x76 [l2cap]
 [<ffffffff8106adea>] ? clockevents_program_event+0x75/0x7e
 [<ffffffff8106bea2>] ? tick_dev_program_event+0x37/0xa5
 [<ffffffffa0394967>] l2cap_sock_release+0x27/0x67 [l2cap]
 [<ffffffff8138c971>] sock_release+0x1a/0x67
 [<ffffffffa03d2492>] rfcomm_session_del+0x34/0x53 [rfcomm]
 [<ffffffffa03d24c5>] rfcomm_session_put+0x14/0x16 [rfcomm]
 [<ffffffffa03d28b4>] rfcomm_session_timeout+0xe/0x1a [rfcomm]
 [<ffffffff810554a8>] run_timer_softirq+0x1e2/0x29a
 [<ffffffffa03d28a6>] ? rfcomm_session_timeout+0x0/0x1a [rfcomm]
 [<ffffffff8104e0f6>] __do_softirq+0xfe/0x1c5
 [<ffffffff8100e8ce>] ? timer_interrupt+0x1a/0x21
 [<ffffffff8100cc4c>] call_softirq+0x1c/0x28
 [<ffffffff8100e05b>] do_softirq+0x33/0x6b
 [<ffffffff8104daf6>] irq_exit+0x36/0x85
 [<ffffffff8100d7a9>] do_IRQ+0xa6/0xbd
 [<ffffffff8100c493>] ret_from_intr+0x0/0xa
 <EOI>  [<ffffffff812585b3>] ? acpi_idle_enter_bm+0x269/0x294
 [<ffffffff812585a9>] ? acpi_idle_enter_bm+0x25f/0x294
 [<ffffffff81373ddc>] ? cpuidle_idle_call+0x97/0x107
 [<ffffffff8100aca0>] ? cpu_idle+0x53/0xaa
 [<ffffffff81429006>] ? rest_init+0x7a/0x7c
 [<ffffffff8177bc8c>] ? start_kernel+0x389/0x394
 [<ffffffff8177b29c>] ? x86_64_start_reservations+0xac/0xb0
 [<ffffffff8177b384>] ? x86_64_start_kernel+0xe4/0xeb

To fix this, the rfcomm_session_put() needs to be moved out of
rfcomm_session_timeout() into rfcomm_process_sessions(). In that
context it is perfectly fine to sleep and disconnect the socket.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Tested-by: David John <davidjon@xenontk.org>
14 years agoBluetooth: Fallback eSCO to SCO on error 0x1a (Unsupported Remote Feature)
Nick Pelly [Wed, 3 Feb 2010 19:42:26 +0000 (11:42 -0800)]
Bluetooth: Fallback eSCO to SCO on error 0x1a (Unsupported Remote Feature)

General Motors carkits that use LGE BT chipsets return this error code
when an eSCO is attempted, despite advertising eSCO support.

2009-08-13 14:41:39.755518 < HCI Command: Setup Synchronous Connection (0x01|0x0028) plen 17
   handle 1 voice setting 0x0060
2009-08-13 14:41:39.757563 > HCI Event: Command Status (0x0f) plen 4
   Setup Synchronous Connection (0x01|0x0028) status 0x00 ncmd 1
2009-08-13 14:41:39.789484 > HCI Event: Synchronous Connect Complete (0x2c) plen 17
   status 0x1a handle 257 bdaddr 00:1E:B2:23:5E:B3 type eSCO
   Error: Unsupported Remote Feature / Unsupported LMP Feature

Signed-off-by: Jaikumar Ganesh <jaikumar@google.com>
Signed-off-by: Nick Pelly <npelly@google.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
14 years agonetfilter: xtables: add CT target
Patrick McHardy [Wed, 3 Feb 2010 16:17:06 +0000 (17:17 +0100)]
netfilter: xtables: add CT target

Add a new target for the raw table, which can be used to specify conntrack
parameters for specific connections, f.i. the conntrack helper.

The target attaches a "template" connection tracking entry to the skb, which
is used by the conntrack core when initializing a new conntrack.

Signed-off-by: Patrick McHardy <kaber@trash.net>
14 years agonetfilter: nf_conntrack: support conntrack templates
Patrick McHardy [Wed, 3 Feb 2010 13:13:03 +0000 (14:13 +0100)]
netfilter: nf_conntrack: support conntrack templates

Support initializing selected parameters of new conntrack entries from a
"conntrack template", which is a specially marked conntrack entry attached
to the skb.

Currently the helper and the event delivery masks can be initialized this
way.

Signed-off-by: Patrick McHardy <kaber@trash.net>
14 years agonetfilter: ctnetlink: support selective event delivery
Patrick McHardy [Wed, 3 Feb 2010 12:51:51 +0000 (13:51 +0100)]
netfilter: ctnetlink: support selective event delivery

Add two masks for conntrack end expectation events to struct nf_conntrack_ecache
and use them to filter events. Their default value is "all events" when the
event sysctl is on and "no events" when it is off. A following patch will add
specific initializations. Expectation events depend on the ecache struct of
their master conntrack.

Signed-off-by: Patrick McHardy <kaber@trash.net>
14 years agonetfilter: nf_conntrack: split up IPCT_STATUS event
Patrick McHardy [Wed, 3 Feb 2010 12:48:53 +0000 (13:48 +0100)]
netfilter: nf_conntrack: split up IPCT_STATUS event

Split up the IPCT_STATUS event into an IPCT_REPLY event, which is generated
when the IPS_SEEN_REPLY bit is set, and an IPCT_ASSURED event, which is
generated when the IPS_ASSURED bit is set.

In combination with a following patch to support selective event delivery,
this can be used for "sparse" conntrack replication: start replicating the
conntrack entry after it reached the ASSURED state and that way it's SYN-flood
resistant.

Signed-off-by: Patrick McHardy <kaber@trash.net>
14 years agonetfilter: add struct net * to target parameters
Patrick McHardy [Wed, 3 Feb 2010 12:45:12 +0000 (13:45 +0100)]
netfilter: add struct net * to target parameters

Signed-off-by: Patrick McHardy <kaber@trash.net>
14 years agonetfilter: ctnetlink: only assign helpers for matching protocols
Patrick McHardy [Wed, 3 Feb 2010 12:41:29 +0000 (13:41 +0100)]
netfilter: ctnetlink: only assign helpers for matching protocols

Make sure not to assign a helper for a different network or transport
layer protocol to a connection.

Additionally change expectation deletion by helper to compare the name
directly - there might be multiple helper registrations using the same
name, currently one of them is chosen in an unpredictable manner and
only those expectations are removed.

Signed-off-by: Patrick McHardy <kaber@trash.net>