summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* net: sock_def_readable() and friends RCU conversionEric Dumazet2010-05-0116-114/+181
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | sk_callback_lock rwlock actually protects sk->sk_sleep pointer, so we need two atomic operations (and associated dirtying) per incoming packet. RCU conversion is pretty much needed : 1) Add a new structure, called "struct socket_wq" to hold all fields that will need rcu_read_lock() protection (currently: a wait_queue_head_t and a struct fasync_struct pointer). [Future patch will add a list anchor for wakeup coalescing] 2) Attach one of such structure to each "struct socket" created in sock_alloc_inode(). 3) Respect RCU grace period when freeing a "struct socket_wq" 4) Change sk_sleep pointer in "struct sock" by sk_wq, pointer to "struct socket_wq" 5) Change sk_sleep() function to use new sk->sk_wq instead of sk->sk_sleep 6) Change sk_has_sleeper() to wq_has_sleeper() that must be used inside a rcu_read_lock() section. 7) Change all sk_has_sleeper() callers to : - Use rcu_read_lock() instead of read_lock(&sk->sk_callback_lock) - Use wq_has_sleeper() to eventually wakeup tasks. - Use rcu_read_unlock() instead of read_unlock(&sk->sk_callback_lock) 8) sock_wake_async() is modified to use rcu protection as well. 9) Exceptions : macvtap, drivers/net/tun.c, af_unix use integrated "struct socket_wq" instead of dynamically allocated ones. They dont need rcu freeing. Some cleanups or followups are probably needed, (possible sk_callback_lock conversion to a spinlock for example...). Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* ipv6: cleanup: remove unneeded null checkDan Carpenter2010-04-301-2/+1
| | | | | | | | | | | | We dereference "sk" unconditionally elsewhere in the function. This was left over from: b30bd282 "ip6_xmit: remove unnecessary NULL ptr check". According to that commit message, "the sk argument to ip6_xmit is never NULL nowadays since the skb->priority assigment expects a valid socket." Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* xfrm: potential uninitialized variable num_xfrmsChangli Gao2010-04-301-1/+1
| | | | | | | | | | | | potential uninitialized variable num_xfrms fix compiler warning: 'num_xfrms' may be used uninitialized in this function. Signed-off-by: Changli Gao <xiaosuo@gmail.com> ---- net/xfrm/xfrm_policy.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) Signed-off-by: David S. Miller <davem@davemloft.net>
* net: speedup sock_recv_ts_and_drops()Eric Dumazet2010-04-302-3/+20
| | | | | | | | | | sock_recv_ts_and_drops() is fat and slow (~ 4% of cpu time on some profiles) We can test all socket flags at once to make fast path fast again. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* cdc_ether: Identify MBM devices by GUID in MDLM descriptorJonas Sjöquist2010-04-301-74/+39
| | | | | | | | | | | | | | | | | | This patch removes vid/pid for Ericsson MBM devices from the whitelist set of devices. The MBM devices are instead identified by GUID. In order for cdc_ether to handle these devices the GUID in the MDLM descriptor is tested. All MBM devices currently handled by cdc_ether as well as future CDC Ethernet MBM devices can be identified by the GUID. This is the same solution used in Carl Nordbeck's mbm driver, http://kerneltrap.org/mailarchive/linux-usb/2008/11/17/4141384/thread I post this as RFC to get feedback on however cdc_ether is the correct place to do the binding, or if it should be done in a separate driver, e.g. zaurus. Signed-off-by: Jonas Sjöquist <jonas.sjoquist@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* forcedeth: Stay in NAPI as long as there's workstephen hemminger2010-04-301-14/+17
| | | | | | | | | | The following does the same thing without the extra overhead of testing all the registers. It also handles the out of memory case. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Tested-by: Tom Herbert <therbert@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* net: ip_queue_rcv_skb() helperEric Dumazet2010-04-287-4/+22
| | | | | | | | | | | | | When queueing a skb to socket, we can immediately release its dst if target socket do not use IP_CMSG_PKTINFO. tcp_data_queue() can drop dst too. This to benefit from a hot cache line and avoid the receiver, possibly on another cpu, to dirty this cache line himself. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* net: speedup udp receive pathEric Dumazet2010-04-284-11/+25
| | | | | | | | | | | | | | | | | | | | | Since commit 95766fff ([UDP]: Add memory accounting.), each received packet needs one extra sock_lock()/sock_release() pair. This added latency because of possible backlog handling. Then later, ticket spinlocks added yet another latency source in case of DDOS. This patch introduces lock_sock_bh() and unlock_sock_bh() synchronization primitives, avoiding one atomic operation and backlog processing. skb_free_datagram_locked() uses them instead of full blown lock_sock()/release_sock(). skb is orphaned inside locked section for proper socket memory reclaim, and finally freed outside of it. UDP receive path now take the socket spinlock only once. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* igb: Clean up left over prototype of igb_get_hw_dev_name()Emil Tantilov2010-04-281-1/+0
| | | | | | Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* wireless: Fix merge.Hauke Mehrtens2010-04-281-1/+0
| | | | | | | | | | in your merge in 5c01d5669356e13f0fb468944c1dd4c6a7e978ad you added "int i;" into wl1271_main.c which is unused in that function. This patch fixes the merge problem: Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de> Signed-off-by: David S. Miller <davem@davemloft.net>
* Bugfix: Link selection was swapped in switch.Sjur Braendeland2010-04-281-2/+2
| | | | | Signed-off-by: Sjur Braendeland <sjur.brandeland@stericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* caif: Bugfixes in CAIF netdevice for close and flow controlSjur Braendeland2010-04-281-57/+73
| | | | | | | | | | | Changes: o Bugfix: Flow control was causing the device to be destroyed. o Bugfix: Handle CAIF channel connect failures. o If the underlying link layer is gone the net-device is no longer removed, but closed. Signed-off-by: Sjur Braendeland <sjur.brandeland@stericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* caif: Rewritten socket implementationSjur Braendeland2010-04-282-969/+831
| | | | | | | | | | | | Changes: This is a complete re-write of the socket layer. Making the socket implementation more aligned with the other socket layers and using more of the support functions available in sock.c. Lots of code is copied from af_unix (and some from af_irda). Non-blocking mode should be working as well. Signed-off-by: Sjur Braendeland <sjur.brandeland@stericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* caif: Disconnect without waiting for responseSjur Braendeland2010-04-283-146/+111
| | | | | | | | | | | | | | | | | | | Changes: o Function cfcnfg_disconn_adapt_layer is changed to do asynchronous disconnect, not waiting for any response from the modem. Due to this the function cfcnfg_linkdestroy_rsp does nothing anymore. o Because disconnect may take down a connection before a connect response is received the function cfcnfg_linkup_rsp is checking if the client is still waiting for the response, if not a disconnect request is sent to the modem. o cfctrl is no longer keeping track of pending disconnect requests. o Added function cfctrl_cancel_req, which is used for deleting a pending connect request if disconnect is done before connect response is received. o Removed unused function cfctrl_insert_req2 o Added better handling of connect reject from modem. Signed-off-by: Sjur Braendeland <sjur.brandeland@stericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* caif: Add reference counting to service layerSjur Braendeland2010-04-287-1/+66
| | | | | | | | | | Changes: o Added functions cfsrvl_get and cfsrvl_put. o Added support release_client to use by socket and net device. o Increase reference counting for in-flight packets from cfmuxl Signed-off-by: Sjur Braendeland <sjur.brandeland@stericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* caif: Rename functions in cfcnfg and caif_devSjur Braendeland2010-04-284-41/+41
| | | | | | | | | | | | Changes: o Renamed cfcnfg_del_adapt_layer to cfcnfg_disconn_adapt_layer o Fixed typo cfcfg to cfcnfg o Renamed linkid to channel_id o Updated documentation in caif_dev.h o Minor formatting changes Signed-off-by: Sjur Braendeland <sjur.brandeland@stericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* caif: Ldisc add permission check and mem-alloc error checkSjur Braendeland2010-04-281-1/+4
| | | | | | | | | | Changes: o Added permission checks for installing. CAP_SYS_ADMIN and CAP_SYS_TTY_CONFIG can install the ldisc. o Check if allocation of skb was successful. Signed-off-by: Sjur Braendeland <sjur.brandeland@stericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Create multiple TX queuesBen Hutchings2010-04-288-123/+140
| | | | | | | | | | | | Create a core TX queue and 2 hardware TX queues for each channel. If separate_tx_channels is set, create equal numbers of RX and TX channels instead. Rewrite the channel and queue iteration macros accordingly. Eliminate efx_channel::used_flags as redundant. Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Test only the first pair of TX queuesBen Hutchings2010-04-282-3/+3
| | | | | | | | | This makes no immediate difference, but we definitely do not want to test all TX queues once we allocate a pair of TX queues to each channel. Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Add Siena PHY BIST and cable diagnostic supportSteve Hodgson2010-04-281-2/+144
| | | | | Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Clean up efx_nic::irq_zero_countBen Hutchings2010-04-281-1/+2
| | | | | | | | There is no need for this to be unsigned long; make it unsigned int. It does need a line in kernel-doc, so add that. Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Add necessary parentheses to macro definitions in net_driver.hBen Hutchings2010-04-281-11/+11
| | | | | Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Break NAPI processing after one ring-full of TX completionsBen Hutchings2010-04-282-24/+33
| | | | | | | | | | | Currently TX completions do not count towards the NAPI budget. This means a continuous stream of TX completions can cause the polling function to loop indefinitely with scheduling disabled. To avoid this, follow the common practice of reporting the budget spent after processing one ring-full of TX completions. Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Set PERIODIC_NOEVENT flag for MC_CMD_MAC_STATSSteve Hodgson2010-04-281-16/+9
| | | | | | | | When set, an event is not sent whenever periodic MAC statistics are raised. This avoids unnecessary wake-ups. Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Update MCDI protocol definitionsBen Hutchings2010-04-281-16/+55
| | | | | Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Enable IPv6 RSS using random key for Toeplitz hashBen Hutchings2010-04-282-0/+21
| | | | | Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Read MEM_STAT for SRM_PERR as well as MEM_PERR errorsSteve Hodgson2010-04-281-1/+2
| | | | | | | | Parity errors in different blocks of SRAM may set one of two different interrupt flags. Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Log specific message for failure of NVRAM self-testBen Hutchings2010-04-281-2/+8
| | | | | Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Extend the legacy interrupt workaroundsSteve Hodgson2010-04-282-15/+10
| | | | | | | | | | | | | | | | | Siena has two problems with legacy interrupts: 1. There is no synchronisation between the ISR read completion, and the interrupt deassert message. 2. A downstream read at the "wrong" moment can return 0, and suppress generating the next interrupt. Falcon should suffer from both of these, and it appears it does. Enable EFX_WORKAROUND_15783 on Falcon as well. Also, when we see queues == 0, ensure we always schedule or rearm every event queue. Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Reconfigure the XAUI serdes after an EM resetSteve Hodgson2010-04-283-1/+5
| | | | | | | | Fix a regression introduced in d3245b28ef2a45ec4e115062a38100bd06229289 "sfc: Refactor link configuration". Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Stop masking out XGMII faults over reconfiguresSteve Hodgson2010-04-281-15/+5
| | | | | | | | | | | | The aim of this code was to avoid a spurious XGMII fault over a MAC reconfigure. It's less relevant now that the PHY reconfigure isn't called from the MAC reconfigure. After applying this patch, our link stress test passed 48 hours of testing without ever resetting the PHY. Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Handle serious errors in exactly one interrupt handlerSteve Hodgson2010-04-283-21/+29
| | | | | | | | | | | | | 'Fatal' errors set an interrupt flag associated with a specific event queue; only read the syndrome vector if we see that queue's flag set (legacy interrupts) or in the interrupt handler for that queue (MSI). Do not ignore an interrupt if the fatal error flag is set but specific error flags are all zero. Even if we don't schedule a reset, we must respect the queue mask and rearm the appropriate event queues. Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Consistently report short MCDI responses as EIOBen Hutchings2010-04-282-11/+17
| | | | | | | | In some cases failing functions were returning 0 which is obviously wrong. In other cases they were returning inappropriate error codes. Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sfc: Ignore parity errors in the other port's SRAMSteve Hodgson2010-04-281-0/+2
| | | | | | | | | | Siena has a separate SRAM bank for each port. On single-port boards these can be merged together, so each port has an interrupt flag for parity errors in the other port's SRAM. Currently we do not enable such merging and should mask this interrupt source. Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* sky2: use the DMA state API instead of the pci equivalentsFUJITA Tomonori2010-04-282-14/+14
| | | | | | | | | | | | | | | This replace the PCI DMA state API (include/linux/pci-dma.h) with the DMA equivalents since the PCI DMA state API will be obsolete. No functional change. For further information about the background: http://marc.info/?l=linux-netdev&m=127037540020276&w=2 Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* bnx2x: Remove two prefetch()Eric Dumazet2010-04-281-2/+0
| | | | | | | | | | 1) Even on 64bit arches, sizeof(struct sk_buff) < 256 2) No need to prefetch same pointer twice. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Acked-by: Eilon Greenstein <eilong@broadcom.com> Acked-by: Eliezer Tamir <eliezer@tamir.org.il> Signed-off-by: David S. Miller <davem@davemloft.net>
* ixgbe: disable MSI-X by default on certain Cisco adaptersNicholas Nunley2010-04-272-1/+20
| | | | | | | | | | | | | Due to an errata in 82598 parts MSI-X needs to be disabled in certain ixgbe devices designed to transfer peer-to-peer traffic on the PCIe bus. This patch sets the default interrupt type to MSI rather than MSI-X for specific Cisco ixgbe adapters. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Acked-by: John Ronciak <john.ronciak@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* bridge: multicast_flood cleanupstephen hemminger2010-04-271-5/+3
| | | | | | | | Move some declarations around to make it clearer which variables are being used inside loop. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* bridge: multicast port group RCU fixstephen hemminger2010-04-272-3/+3
| | | | | | | | | | | | The recently introduced bridge mulitcast port group list was only partially using RCU correctly. It was missing rcu_dereference() and missing the necessary barrier on deletion. The code should have used one of the standard list methods (list or hlist) instead of open coding a RCU based link list. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* bridge: multicast floodstephen hemminger2010-04-271-2/+2
| | | | | | | | Fix unsafe usage of RCU. Would never work on Alpha SMP because of lack of rcu_dereference() Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* bridge: simplify multicast_add_routerstephen hemminger2010-04-271-8/+11
| | | | | | | | By coding slightly differently, there are only two cases to deal with: add at head and add after previous entry. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* ixgbe: add registers etc. printout code just before resetting adaptersTaku Izumi2010-04-271-0/+341
| | | | | | | | | | | This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* igb: add registers etc. printout code just before resetting adaptersTaku Izumi2010-04-271-0/+332
| | | | | | | | | | | This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* e1000e: add registers etc. printout code just before resetting adaptersTaku Izumi2010-04-271-0/+357
| | | | | | | | | | | This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* e1000: Use netdev_<level>, pr_<level> and dev_<level>Emil Tantilov2010-04-277-405/+367
| | | | | | | | | | | | | | | | | This patch is an alternative to similar patch provided by Joe Perches. Substitute DPRINTK macro for e_<level> that uses netdev_<level> and dev_<level> similar to e1000e. - Convert printk to pr_<level> where applicable. - Use common #define pr_fmt for the driver. - Use dev_<level> for displaying text in parts of the driver where the interface name is not assigned (like e1000_param.c). - Better align test with the new macros. CC: Joe Perches <joe@perches.com> Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* Revert "bridge: Use hlist_for_each_entry_rcu() in br_multicast_add_router()"David S. Miller2010-04-271-1/+1
| | | | | | | | | | | This reverts commit ff65e8275f6c96a5eda57493bd84c4555decf7b3. As explained by Stephen Hemminger, the traversal doesn't require RCU handling as we hold a lock. The list addition et al. calls, on the other hand, do. Signed-off-by: David S. Miller <davem@davemloft.net>
* ixgbevf: use DMA API instead of PCI DMA functionsNick Nunley2010-04-271-33/+35
| | | | | | 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>
* ixgbe: use DMA API instead of PCI DMA functionsNick Nunley2010-04-272-55/+64
| | | | | | 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>
* ixgb: use DMA API instead of PCI DMA functionsNick Nunley2010-04-271-31/+39
| | | | | | 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>
* igbvf: use DMA API instead of PCI DMA functionsNick Nunley2010-04-271-36/+38
| | | | | | 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>
OpenPOWER on IntegriCloud