summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2009-11-02ipv6: no more dev_put() in inet6_bind()Eric Dumazet1-8/+9
Avoids touching device refcount in inet6_bind(), thanks to RCU Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-02ip6tnl: less dev_put() callsEric Dumazet1-6/+5
Using dev_get_by_index_rcu() in ip6_tnl_rcv_ctl() & ip6_tnl_xmit_ctl() avoids touching device refcount. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-02packet: less dev_put() callsEric Dumazet1-10/+12
- packet_sendmsg_spkt() can use dev_get_by_name_rcu() to avoid touching device refcount. - packet_getname_spkt() & packet_getname() can use dev_get_by_index_rcu() to avoid touching device refcount too. tpacket_snd() & packet_snd() can not use RCU yet because they can sleep when allocating skb. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-01pppoe: RCU locking in get_item_by_addr()Eric Dumazet1-10/+9
Use dev_get_by_name_rcu() instead of dev_get_by_name(), to avoid touching device refcount in hotpath. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-01ifb: RCU locking avoids touching dev refcountEric Dumazet1-2/+4
Avoids touching dev refcount in hotpath Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-01net: RCU locking for simple ioctl()Eric Dumazet1-4/+4
All ioctls() implemented by dev_ifsioc_locked() : SIOCGIFFLAGS, SIOCGIFMETRIC, SIOCGIFMTU, SIOCGIFHWADDR, SIOCGIFSLAVE, SIOCGIFMAP, SIOCGIFINDEX & SIOCGIFTXQLEN can use RCU lock instead of dev_base_lock rwlock Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-01icmp: icmp_send() can avoid a dev_put()Eric Dumazet1-4/+5
We can avoid touching device refcount in icmp_send(), using dev_get_by_index_rcu() Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-01ipv4: inetdev_by_index() switch to RCUEric Dumazet1-3/+4
Use dev_get_by_index_rcu() instead of __dev_get_by_index() and dev_base_lock rwlock Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-01veth: Fix unregister_netdevice_queue for vethEric W. Biederman1-1/+2
I tested the recent unregister many changes and got a weird, nasty and seemingly unrelasted kernel oops. Changing unregister_netdevice_queue to use list_move_tail fixes the problem for me. ip link add type veth rmmod veth ls /sys/class/net/ showed one of the veth devices still present. A subsequent ip link oopsed the box. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-01net: Introduce dev_get_by_name_rcu()Eric Dumazet2-9/+41
Some workloads hit dev_base_lock rwlock pretty hard. We can use RCU lookups to avoid touching this rwlock (and avoid touching netdevice refcount) netdevices are already freed after a RCU grace period, so this patch adds no penalty at device dismantle time. However, it adds a synchronize_rcu() call in dev_change_name() Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-01NET:KS8695: add API for get rx interrupt bitFigo.zhang1-12/+55
1. Add API Add k8695_get_rx_enable_bit() for get Rx interrupt enable/status bit. 2. add some comment or document about some functions and variables. 3. update driver version to "1.02" 4. add napi_enable() and napi_disable() in open/close file method. Signed-off-by: Figo.zhang <figo1802@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-30RDS/IB+IW: Move recv processing to a taskletAndy Grover6-12/+52
Move receive processing from event handler to a tasklet. This should help prevent hangcheck timer from going off when RDS is under heavy load. Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-30RDS: Do not send congestion updates to loopback connectionsAndy Grover1-0/+2
This issue was discovered by HP's Pradeep and fixed in OFED 1.3, but not fixed in later versions, since the fix's implementation was not immediately applyable to the later code. This patch should do the trick for 1.4+ codebases. Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-30RDS: Fix panic on unloadAndy Grover2-8/+2
Remove explicit destruction of passive connection when destroying active end of the connection. The passive end is also on the device's connection list, and will thus be cleaned up properly. Panic was caused by trying to clean it up twice. Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-30RDS: Fix potential race around rds_i[bw]_allocationAndy Grover2-6/+8
"At rds_ib_recv_refill_one(), it first executes atomic_read(&rds_ib_allocation) for if-condition checking, and then executes atomic_inc(&rds_ib_allocation) if the condition was not satisfied. However, if any other code which updates rds_ib_allocation executes between these two atomic operation executions, it seems that it may result race condition. (especially when rds_ib_allocation + 1 == rds_ib_sysctl_max_recv_allocation)" This patch fixes this by using atomic_inc_unless to eliminate the possibility of allocating more than rds_ib_sysctl_max_recv_allocation and then decrementing the count if the allocation fails. It also makes an identical change to the iwarp transport. Reported-by: Shin Hong <hongshin@gmail.com> Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-30RDS: Add GET_MR_FOR_DEST sockoptAndy Grover4-0/+36
RDS currently supports a GET_MR sockopt to establish a memory region (MR) for a chunk of memory. However, the fastreg method ties a MR to a particular destination. The GET_MR_FOR_DEST sockopt allows the remote machine to be specified, and thus support for fastreg (aka FRWRs). Note that this patch does *not* do all of this - it simply implements the new sockopt in terms of the old one, so applications can begin to use the new sockopt in preparation for cutover to FRWRs. Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-30bond: Get the rtnl_link_ops support correctEric W. Biederman1-1/+3
- Don't call rtnl_link_unregister if rtnl_link_register fails - Set .priv_size so we aren't stomping on uninitialized memory when we use netdev_priv, on bond devices created with ip link add type bond. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-30bond: Add support for multiple network namespacesEric W. Biederman6-58/+99
Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-30bond: Implement a basic set of rtnl link opsEric W. Biederman1-25/+30
This implements a basic set of rtnl link ops and takes advantage of the fact that rtnl_link_unregister kills all of the surviving devices to all us to kill bond_free_all. A module alias is added so ip link add can pull in the bonding module. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-30bond: Simplify bond device destructionEric W. Biederman1-29/+16
Manually inline the code from bond_deinit to bond_uninit. bond_uninit is the only caller and it is short. Move the call of bond_release_all from the netdev notifier into bond_uninit. The call site is effectively the same and performing the call explicitly allows all the paths for destroying a bonding device to behave the same way. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-30bond: Simplify bond_create.Eric W. Biederman1-18/+4
Stop calling dev_get_by_name to see if the bond device already exists. register_netdevice already does that. Stop calling bond_deinit if register_netdevice fails as bond_uninit is guaranteed to be called if bond_init succeeds. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-30bond: Simply bond sysfs group creationEric W. Biederman3-30/+4
This patch delegates the work of creating the sysfs groups to the netdev layer and ultimately to the device layer. This closes races between uevents. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-30net: Allow devices to specify a device specific sysfs group.Eric W. Biederman2-3/+6
This isn't beautifully abstracted, but it is simple, simplifies uses and so far is only needed for the bonding driver. Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-30net: use hlist_for_each_entry()Eric Dumazet1-8/+8
Small cleanup of __dev_get_by_name() and __dev_get_by_index() to use hlist_for_each_entry() : They'll look like their _rcu variant. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-30veth: Fix veth_dellink methodEric Dumazet1-2/+2
In commit 23289a37e2b127dfc4de1313fba15bb4c9f0cd5b (net: add a list_head parameter to dellink() method), I forgot to actually use this parameter in veth_dellink. I remember feeling a bit uncomfortable about veth_close(), because it does : netif_carrier_off(dev); netif_carrier_off(priv->peer); Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29vlan: cleanup multiple unregistrationsPatrick McHardy1-32/+20
The temporary copy of the VLAN group is not neccessary since the lower device is already in the process of being unregistered, if it was neccessary the memset of the global group would introduce a race condition. With this removed, the changes to the original code are only a few lines, so remove the new function and move the code back into vlan_device_event(). Signed-off-by: Patrick McHardy <kaber@trash.net> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29convert kaweth to use usb_reset_configuration()Oliver Neukum1-10/+1
For USB 3.0 it is necessary that all drivers use the standard API to reset a configuration. This removes a home-grown implementation. Signed-off-by: Oliver Neukum <oliver@neukum.org> Hi David, please take this for the next merge window. Regards Oliver Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29ax25: unsigned cannot be less than 0 in ax25_ctl_ioctl()roel kluin1-4/+3
struct ax25_ctl_struct member `arg' is unsigned and cannot be less than 0. Signed-off-by: Roel Kluin <roel.kluin@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29sky2: version 1.26Stephen Hemminger1-1/+1
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29sky2: 88E8059 supportStephen Hemminger1-1/+78
Tentative support for newer Marvell hardware including the Yukon-2 Optima chip. Do not have hatdware to test this yet, code is based on vendor driver. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29sky2: workarounds for Yukon-2 supremeStephen Hemminger1-2/+32
Changes related to support of Yukon supreme chip. Don't have this chip version to test on, these are reverse engineered from the vendor (GPL) driver. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29sky2: fix receive pause thresholdsStephen Hemminger2-5/+12
Program the receive pause thresholds differently depending on chip version. This cloned from from the vendor (GPL) driver. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29sky2: add register definitions for new chipsStephen Hemminger1-1/+177
This adds infrastructure for the newer chip versions and workarounds. Extracted from the vendor (GPL) driver. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29sky2: add SK-9E21M device idStephen Hemminger1-0/+1
This is a new ID that just showed up in latest vendor driver. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29sfc: Enable heuristic selection between page and skb RX buffersBen Hutchings1-1/+1
Now that we can tell whether GRO is being applied, this heuristic is effective once more. Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29sfc: Feed GRO result into RX allocation policy and interrupt moderationBen Hutchings1-2/+11
When GRO is successfully merging received packets, we should allocate raw page buffers rather than skbs that will be discarded by GRO. Otherwise, we should allocate skbs. GRO also benefits from higher interrupt moderation, so increase the score for mergeable RX packets. Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29gro: Change all receive functions to return GRO result codesBen Hutchings4-45/+42
This will allow drivers to adjust their receive path dynamically based on whether GRO is being applied successfully. Currently all in-tree callers ignore the return values of these functions and do not need to be changed. Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Acked-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29gro: Name the GRO result enumeration typeBen Hutchings3-11/+23
This clarifies which return and parameter types are GRO result codes and not RX result codes. Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Acked-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29Merge branch 'master' of ↵David S. Miller25-83/+160
master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
2009-10-29net: Fix 'Re: PACKET_TX_RING: packet size is too long'Gabor Gombas1-4/+1
Currently PACKET_TX_RING forces certain amount of every frame to remain unused. This probably originates from an early version of the PACKET_TX_RING patch that in fact used the extra space when the (since removed) CONFIG_PACKET_MMAP_ZERO_COPY option was enabled. The current code does not make any use of this extra space. This patch removes the extra space reservation and lets userspace make use of the full frame size. Signed-off-by: Gabor Gombas <gombasg@sztaki.hu> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29net,socket: introduce DECLARE_SOCKADDR helper to catch overflow at build timeCyrill Gorcunov3-1/+7
proto_ops->getname implies copying protocol specific data into storage unit (particulary to __kernel_sockaddr_storage). So when we implement new protocol support we should keep such a detail in mind (which is easy to forget about). Lets introduce DECLARE_SOCKADDR helper which check if storage unit is not overfowed at build time. Eventually inet_getname is switched to use DECLARE_SOCKADDR (to show example of usage). Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29Merge branch 'master' of ↵David S. Miller137-6003/+13469
git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-next-2.6
2009-10-29net: Introduce dev_get_by_index_rcu()Eric Dumazet2-10/+39
Some workloads hit dev_base_lock rwlock pretty hard. We can use RCU lookups to avoid touching this rwlock. netdevices are already freed after a RCU grace period, so this patch adds no penalty at device dismantle time. dev_ifname() converted to dev_get_by_index_rcu() Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29atm: Cleanup redundant tests on unsignedroel kluin2-3/+3
The variables are unsigned so the `< 0' test always fails, the other part of the test catches wrapped values. Signed-off-by: Roel Kluin <roel.kluin@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29net: Cleanup redundant tests on unsignedroel kluin3-6/+1
optlen is unsigned so the `< 0' test is never true. Signed-off-by: Roel Kluin <roel.kluin@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29net: Cleanup redundant tests on unsignedroel kluin2-2/+2
If there is data, the unsigned skb->len is greater than 0. rt.sigdigits is unsigned as well, so the test `>= 0' is always true, the other part of the test catches wrapped values. Signed-off-by: Roel Kluin <roel.kluin@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29gigaset: convert strcmp chain to table lookupTilman Schmidt2-23/+25
Replace the sequence of strcmp calls for interpreting ZSAU parameter strings by a table of known strings and lookup loop to improve readability. Impact: readability improvement, no functional change Signed-off-by: Tilman Schmidt <tilman@imap.cc> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29gigaset: checkpatch cleanup of ev-layer.cTilman Schmidt1-247/+231
On more step towards the holy grail of checkpatch.pl silence. Impact: cosmetic Signed-off-by: Tilman Schmidt <tilman@imap.cc> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29bas_gigaset: checkpatch cleanupTilman Schmidt2-82/+93
On the quest for the holy grail of checkpatch.pl silence. Impact: cosmetic Signed-off-by: Tilman Schmidt <tilman@imap.cc> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-29ser_gigaset: checkpatch cleanupTilman Schmidt1-19/+37
Duly uglified as demanded by checkpatch.pl. Impact: cosmetic Signed-off-by: Tilman Schmidt <tilman@imap.cc> Signed-off-by: David S. Miller <davem@davemloft.net>
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-06 03:38:56 +0000