Commit Graph

168332 Commits

Author SHA1 Message Date
Eric Dumazet
e014debecd linkwatch: linkwatch_forget_dev() to speedup device dismantle
Herbert Xu a écrit :
> On Tue, Nov 17, 2009 at 04:26:04AM -0800, David Miller wrote:
>> Really, the link watch stuff is just due for a redesign.  I don't
>> think a simple hack is going to cut it this time, sorry Eric :-)
>
> I have no objections against any redesigns, but since the only
> caller of linkwatch_forget_dev runs in process context with the
> RTNL, it could also legally emit those events.

Thanks guys, here an updated version then, before linkwatch surgery ?

In this version, I force the event to be sent synchronously.

[PATCH net-next-2.6] linkwatch: linkwatch_forget_dev() to speedup device dismantle

time ip link del eth3.103 ; time ip link del eth3.104 ; time ip link del eth3.105

real	0m0.266s
user	0m0.000s
sys	0m0.001s

real	0m0.770s
user	0m0.000s
sys	0m0.000s

real	0m1.022s
user	0m0.000s
sys	0m0.000s

One problem of current schem in vlan dismantle phase is the
holding of device done by following chain :

vlan_dev_stop() ->
	netif_carrier_off(dev) ->
		linkwatch_fire_event(dev) ->
			dev_hold() ...

And __linkwatch_run_queue() runs up to one second later...

A generic fix to this problem is to add a linkwatch_forget_dev() method
to unlink the device from the list of watched devices.

dev->link_watch_next becomes dev->link_watch_list (and use a bit more memory),
to be able to unlink device in O(1).

After patch :
time ip link del eth3.103 ; time ip link del eth3.104 ; time ip link del eth3.105

real    0m0.024s
user    0m0.000s
sys     0m0.000s

real    0m0.032s
user    0m0.000s
sys     0m0.001s

real    0m0.033s
user    0m0.000s
sys     0m0.000s

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-18 05:03:11 -08:00
Octavian Purdila
e2ce146848 ipv4: factorize cache clearing for batched unregister operations
Signed-off-by: Octavian Purdila <opurdila@ixiacom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-18 05:03:07 -08:00
Octavian Purdila
395264d509 net: introduce NETDEV_UNREGISTER_PERNET
This new event is called once for each unique net namespace in batched
unregister operations (with the argument set to a random device from
that namespace) and once per device in non-batched unregister
operations.

It allows us to factorize some device unregister work such as clearing the
routing cache.

Signed-off-by: Octavian Purdila <opurdila@ixiacom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-18 05:03:03 -08:00
Eric Dumazet
fccaf71011 macvlan: Precise RX stats accounting
With multi queue devices, its possible that several cpus call
macvlan RX routines simultaneously for the same macvlan device.

We update RX stats counter without any locking, so we can
get slightly wrong counters.

One possible fix is to use percpu counters, to get precise
accounting and also get guarantee of no cache line ping pongs
between cpus.

Note: this adds 16 bytes (32 bytes on 64bit arches) of percpu
data per macvlan device.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 23:51:57 -08:00
Eric Dumazet
9793241fe9 vlan: Precise RX stats accounting
With multi queue devices, its possible that several cpus call
vlan RX routines simultaneously for the same vlan device.

We update RX stats counter without any locking, so we can
get slightly wrong counters.

One possible fix is to use percpu counters, to get precise
accounting and also get guarantee of no cache line ping pongs
between cpus.

Note: this adds 16 bytes (32 bytes on 64bit arches) of percpu
data per vlan device.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 23:51:55 -08:00
Eric Dumazet
d83345adf9 net: add dev_txq_stats_fold() helper
Some drivers ndo_get_stats() method need to perform txqueue stats folding.

Move folding from dev_get_stats() to a new dev_txq_stats_fold() function

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 23:51:52 -08:00
Stephen Rothwell
b038b0401f vmxnet3: using csum_ipv6_magic requires including net/ip6_checksum.h
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 23:04:59 -08:00
Heiko Carstens
302689ac47 net/s390 drivers: add missing 'const' attribute
Add missing 'const' attribute to avoid the following compile warnings:

drivers/s390/net/ctcm_main.c: In function 'ctcm_init':
drivers/s390/net/ctcm_main.c:1864: warning: assignment from incompatible pointer type
drivers/s390/net/lcs.c: In function 'lcs_init_module':
drivers/s390/net/lcs.c:2468: warning: assignment from incompatible pointer type
drivers/s390/net/claw.c: In function 'claw_init':
drivers/s390/net/claw.c:3408: warning: assignment from incompatible pointer type

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 06:47:02 -08:00
Changli Gao
b76965e02b act_mirred: optimization.
move checking if eaction is valid in tcf_mirred_init()

Signed-off-by: Changli Gao <xiaosuo@gmail.com>
Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 04:15:38 -08:00
Changli Gao
feed1f1724 act_mirred: cleanup
1. don't let go back using goto.
2. don't call skb_act_clone() until it is necessary.
3. one exit of the critical context.

Signed-off-by: Changli Gao <xiaosuo@gmail.com>
Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 04:15:37 -08:00
Rémi Denis-Courmont
b2a5decddb Phonet: missing rcu_dereference()
Reported-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 04:08:50 -08:00
Shreyas Bhatewara
115924b6bd net: Getting rid of the x86 dependency to built vmxnet3
This patch removes config dependency on x86 to build vmxnet3 driver. Thus
the driver can be built on big endian architectures now. Although vmxnet3
is not supported on VMs other than x86 architecture, all this code goes in
to ensure correctness. If the code is not dependent on x86, it should not
assume little endian architecture in any of its operations.

Signed-off-by: Shreyas Bhatewara <sbhatewara@vmware.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 04:08:50 -08:00
Johannes Berg
649300b927 netlink: remove subscriptions check on notifier
The netlink URELEASE notifier doesn't notify for
sockets that have been used to receive multicast
but it should be called for such sockets as well
since they might _also_ be used for sending and
not solely for receiving multicast. We will need
that for nl80211 (generic netlink sockets) in the
future.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Cc: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 04:08:49 -08:00
Vladislav Zolotarov
54b9ddaa68 bnx2x: Handle Rx and Tx together in NAPI
Put Tx and Rx DPC to be handled in the NAPI:
  - Saves status blocks.
  - Moves the Tx work from hardIRQ to NAPI.

Signed-off-by: Vladislav Zolotarov <vladz@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 04:08:48 -08:00
fangxiaozhi
208f2037ae net: PPP buffer too small for higher speed connections
1. This patch is based on the kernel of 2.6.32-rc7 

2. In this patch, we enlarge the out buffer size to optimize the
   upload speed for the ppp connection. Then it can support the upload of
   HSUPA data cards.

Signed-off-by: fangxiaozhi <huananhu@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 04:02:24 -08:00
Wolfram Sang
24cfbcbabf net/can/mscan: improve build
- move Kconfig entries to the subdirectory
- do remaining renames of mpc52xx to mpc5xxx

Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 03:57:34 -08:00
Wolfram Sang
323907ac72 net/can/mscan: add error path to mscan_open()
Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 03:54:05 -08:00
Wolfram Sang
74ff60b29e net/can/mscan: replace hardcoded values with defines
Not all hardcoded values have been replaced as this made the code quite
unreadable. IMHO this compromise serves the purpose of readability.

Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 03:54:04 -08:00
Wolfram Sang
81593c1cea net/can/mpc52xx_can: improve properties and their description
Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Cc: devicetree-discuss@ozlabs.org
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 03:54:04 -08:00
Wolfram Sang
3f158c2532 net/can/mpc52xx_can: refactor clock-get routine
Merge two functions into one. The result is smaller as they can now share some
variables.

Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 03:54:03 -08:00
Wolfram Sang
68bd7422a2 net/can/mscan: drop assignment in while-construct
As suggested by Wolfgang Grandegger.

Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 03:54:02 -08:00
Wolfram Sang
1712fe5941 net/can/mscan: fix function annotations
- use extern where apropriate
- don't export symbols

Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 03:54:02 -08:00
Wolfram Sang
59179ea60c net/can/mscan: use {clr|set}bits8 macros
Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 03:54:01 -08:00
Wolfram Sang
377a1f0b92 net/can/mscan: drop support for CAN_MODE_{SLEEP|STOP}
The upper layer does not support it yet.

Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 03:54:00 -08:00
Wolfram Sang
0285e7ceaa net/can/mscan: trivial fixes
- remove whitespaces
- use ! and ?: when apropriate
- make braces consistent

Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 03:54:00 -08:00
Wolfram Sang
622ed7e9cf net/can/mscan: move defines into .h file
Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-17 03:53:59 -08:00
David S. Miller
a2bfbc072e Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
Conflicts:
	drivers/net/can/Kconfig
2009-11-17 00:05:02 -08:00
zeal
5c427ff9e4 KS8695: fix ks8695_rx() unreasonable action.
ks8695_rx() will call refill_buffers() for every incoming packet.
Its not necessary. We just need do it after finishing receiving thing.
And the 'RX dma engine' is in the same situation.
This blocks our user space application. The following patch may fix it.

Signed-off-by: zeal <zealcook@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-16 23:51:47 -08:00
zeal
fa6cae143d KS8695: fix ks8695_rx_irq() bug.
ks8695 rx irq is edge-level. Before arriving at irq handler, the
corresponding status bit has been clear(irq's ack).
So we should not check it after that.

Signed-off-by: zeal <zealcook@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-16 23:51:47 -08:00
Ben McKeegan
82b3cc1a2f ppp: fix BUG on non-linear SKB (multilink receive)
PPP does not correctly call pskb_may_pull() on all necessary receive paths
before reading the PPP protocol, thus causing PPP to report seemingly
random 'unsupported protocols' and eventually trigger BUG_ON(skb->len <
skb->data_len) in skb_pull_rcsum() when receiving multilink protocol in
non-linear skbs.

ppp_receive_nonmp_frame() does not call pskb_may_pull() before reading the
protocol number.  For the non-mp receive path this is not a problem, as
this check is done in ppp_receive_frame().  For the mp receive path,
ppp_mp_reconstruct() usually copies the data into a new linear skb.
However, in the case where the frame is made up of a single mp fragment,
the mp header is pulled and the existing skb used.  This skb was then
passed to ppp_receive_nonmp_frame() without checking if the encapsulated
protocol header could safely be read.

Signed-off-by: Ben McKeegan <ben@netservers.co.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-16 23:51:34 -08:00
Breno Leitao
c0e1f68bce ixgbe: Fixing EEH handler to handle more than one error
After commmit 4b77b0a2ba EEH breaks
after the second error, since it calls pci_restore_state()
but it returns 0, since pci->state_saved is false.

So, this patch just call pci_save_state() after pci_restore_state().

Signed-off-by: Breno Leitao <leitao@linux.vnet.ibm.com>
Acked-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-16 23:51:32 -08:00
David S. Miller
8494347eed Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-2.6 2009-11-16 23:43:01 -08:00
Eric Dumazet
91e9c07bd6 net: Fix the rollback test in dev_change_name()
net: Fix the rollback test in dev_change_name()

In dev_change_name() an err variable is used for storing the original
call_netdevice_notifiers() errno (negative) and testing for a rollback
error later, but the test for non-zero is wrong, because the err might
have positive value as well - from dev_alloc_name(). It means the
rollback for a netdevice with a number > 0 will never happen. (The err
test is reordered btw. to make it more readable.)

Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-16 03:30:35 -08:00
Frank Blaschka
3fd434d846 qeth: allow dynamic change of rx checksumming
Technically there is no need to set the card offline to change
RX checksumming. Get rid of this stupid limitation.

Signed-off-by: Frank Blaschka <frank.blaschka@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-16 02:42:09 -08:00
Frank Blaschka
c3b4a740db qeth: rework TSO functions
The maximum TSO size OSA can handle is 15 * PAGE_SIZE. This
patch reduces gso_max_size to this value and adds some sanity
checks and statistics to the TSO implementation.
Since only layer 3 is able to do TSO move all TSO related functions
to the qeth_l3 module.

Signed-off-by: Frank Blaschka <frank.blaschka@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-16 02:42:08 -08:00
Ursula Braun
aa90922479 qeth: Recognize return codes of ccw_device_set_online
Setting a qeth device online requires to call function
ccw_device_set_online() for read-, write-, and data-subchannel.
Failures should be detected immediately without an attempt to
invoke follow-on activity qeth_qdio_clear_card().,

In addition, ccw_device_set_online calls are consolidated in
qeth_core_main.c only.

Signed-off-by: Ursula Braun <ursula.braun@de.ibm.com>
Signed-off-by: Frank Blaschka <frank.blaschka@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-16 02:42:07 -08:00
Ursula Braun
f20b04597b qeth: remaining EDDP cleanup
EDDP code has been removed from qeth in 2009. This patch removes two
useless remaining EDDP-references.

Signed-off-by: Ursula Braun <ursula.braun@de.ibm.com>
Signed-off-by: Frank Blaschka <frank.blaschka@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-16 02:42:06 -08:00
Einar Lueck
d64ecc22d0 qeth: Exploit Connection Isolation
Isolate data connection to a shared OSA card against other data
connections to the same OSA card. Connectivity between isolated
data connections sharing the same OSA card is therefore possible only
through external network gear (e.g. a router).

Signed-off-by: Einar Lueck <elelueck@de.ibm.com>
Signed-off-by: Frank Blaschka <frank.blaschka@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-16 02:42:05 -08:00
David S. Miller
e29d436317 Revert "isdn: isdn_ppp: Use SKB list facilities instead of home-grown implementation."
This reverts commit 38783e6713.

It causes kernel bugzilla #14594

Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-15 22:23:47 -08:00
Marin Mitov
b9f5d52670 remove deprecated and not used: print_mac()
The function print_mac in net/ethernet/eth.c is marked __deprecated
and not used. Remove it.

Signed-off-by: Marin Mitov <mitov@issp.bas.bg>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-15 22:21:34 -08:00
Jay Vosburgh
2d6682db11 bonding: fix 802.3ad standards compliance error
The language of 802.3ad 43.4.9 requires the "recordPDU" function
to, in part, compare the Partner parameter values in a received LACPDU
to the stored Actor values.  If those match, then the Partner's
synchronization state is set to true.

	The current 802.3ad implementation is performing these steps out
of order; first, the synchronization check is done, then the paramters are
checked to see if they match (the synch check being done against a match
check of a prior LACPDU).  This causes delays in establishing aggregators
in some circumstances.

	This patch modifies the 802.3ad code to call __choose_matched,
the function that does the "match" comparisions, as the first step of
__record_pdu, instead of immediately afterwards.  This new behavior is
in compliance with the language of the standard.

	Some additional commentary relating to code vs. standard is also
added.

	Reported by Martin Patterson <martin@gear6.com> who also supplied
the logic of the fix and verified the patch.

Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-15 22:21:34 -08:00
Eric Dumazet
b93ab837a2 vlan: Use __vlan_hwaccel_put_tag() in rx
Commit 05423b2413 (vlan: allow null VLAN ID to be used)
forgot to update __vlan_hwaccel_rx() & vlan_gro_common()

We need to set VLAN_TAG_PRESENT flag in skb->vlan_tci

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-15 22:21:33 -08:00
Sriram
5bfa2a17f2 TI Davinci EMAC : Fix Console Hang when bringing the interface down
In the NAPI poll function(emac_poll), check for netif_running()
is unnecassary. In addition to associated runtime overhead, it
also results in a continuous softirq loop when the interface is
brought down under heavy traffic(tested wit Traffic Generator).
Once the interface is disabled, the poll function always returns
zero(with the check for netif_running) and napi_complete() would
never get called resulting in softirq loop.

Signed-off-by: Sriramakrishnan <srk@ti.com>
Acked-by: Chaithrika U S <chaithrika@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-15 22:17:43 -08:00
Sriram
f88c5b9887 smsc911x: Fix Console Hang when bringing the interface down.
In the NAPI poll function, check for netif_running() is unnecassary.
In addition to associated runtime overhead, it also results in
continuous softirq loop when the interface is brought down under heavy
traffic(tested with Traffic Generator).Once the interface is disabled,
the poll function always returns zero(with the check for netif_running)
and napi_complete() would never get called resulting in softirq loop.

Signed-off-by: Sriramakrishnan <srk@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-15 22:17:41 -08:00
Matt Carlson
c5d5d17217 tg3: Update version to 3.104
This patch updates the tg3 version to 3.104.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-15 22:14:55 -08:00
Matt Carlson
5001e2f638 tg3: Fix DIDs, Enable 5717 support
This patch fixes the 5717 variant device ID enumerations and adds those
DIDs to the PCI ID table.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-15 22:14:54 -08:00
Matt Carlson
b196c7e45f tg3: Add rx prod ring consolidation
This patch adds code to funnel each MSI-X vector's rx packet buffers
into a single set of producer rings which will then be submitted to the
hardware.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-15 22:14:54 -08:00
Matt Carlson
66711e6663 tg3: Create aliases for rx producer mailbox regs
The rx producer mailbox registers are used in several spots in the code.
The addition of TG3_64BIT_REG_LOW makes register references
uncomfortably long.  This patch creates an alias for the standard and
jumbo ring producer index registers to make the code cleaner.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-15 22:14:53 -08:00
Matt Carlson
2b2cdb65be tg3: Lay proucer ring handling groundwork
The patch increases the number of producer rings available and
implements the constructor and destructor code that deals with them.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-15 22:14:51 -08:00
Matt Carlson
4361935afe tg3: Consider rx_std_prod_idx a hw mailbox
This patch changes how the code uses the rx_std_prod_idx member.  In the
following patch, the code will be changed so that it will act just like
a hardware mailbox.  This patch prepares the code so that memory barriers
can be more easily inserted.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-15 22:14:50 -08:00