aboutsummaryrefslogtreecommitdiffstats
path: root/include/net
Commit message (Collapse)AuthorAge
...
* | sctp: Send user messages to the lower layer as oneVlad Yasevich2009-09-04
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currenlty, sctp breaks up user messages into fragments and sends each fragment to the lower layer by itself. This means that for each fragment we go all the way down the stack and back up. This also discourages bundling of multiple fragments when they can fit into a sigle packet (ex: due to user setting a low fragmentation threashold). We introduce a new command SCTP_CMD_SND_MSG and hand the whole message down state machine. The state machine and the side-effect parser will cork the queue, add all chunks from the message to the queue, and then un-cork the queue thus causing the chunks to get transmitted. Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
* | sctp: Disallow new connection on a closing socketVlad Yasevich2009-09-04
| | | | | | | | | | | | | | | | | | | | | | If a socket has a lot of association that are in the process of of being closed/aborted, it is possible for a remote to establish new associations during the time period that the old ones are shutting down. If this was a result of a close() call, there will be no socket and will cause a memory leak. We'll prevent this by setting the socket state to CLOSING and disallow new associations when in this state. Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
* | sctp: remove unused union (sctp_cmsg_data_t) definitionRami Rosen2009-09-04
| | | | | | | | | | | | | | | | This patch removes an unused union definition (sctp_cmsg_data_t) from include/net/sctp/user.h. Signed-off-by: Rami Rosen <rosenrami@gmail.com> Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
* | tcp: replace hard coded GFP_KERNEL with sk_allocationWu Fengguang2009-09-03
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This fixed a lockdep warning which appeared when doing stress memory tests over NFS: inconsistent {RECLAIM_FS-ON-W} -> {IN-RECLAIM_FS-W} usage. page reclaim => nfs_writepage => tcp_sendmsg => lock sk_lock mount_root => nfs_root_data => tcp_close => lock sk_lock => tcp_send_fin => alloc_skb_fclone => page reclaim David raised a concern that if the allocation fails in tcp_send_fin(), and it's GFP_ATOMIC, we are going to yield() (which sleeps) and loop endlessly waiting for the allocation to succeed. But fact is, the original GFP_KERNEL also sleeps. GFP_ATOMIC+yield() looks weird, but it is no worse the implicit sleep inside GFP_KERNEL. Both could loop endlessly under memory pressure. CC: Arnaldo Carvalho de Melo <acme@ghostprotocols.net> CC: David S. Miller <davem@davemloft.net> CC: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | vlan: multiqueue vlan deviceEric Dumazet2009-09-02
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | vlan devices are currently not multi-queue capable. We can do that with a new rtnl_link_ops method, get_tx_queues(), called from rtnl_create_link() This new method gets num_tx_queues/real_num_tx_queues from real device. register_vlan_device() is also handled. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | inet: inet_connection_sock_af_ops constStephen Hemminger2009-09-02
| | | | | | | | | | | | | | | | The function block inet_connect_sock_af_ops contains no data make it constant. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | Merge branch 'master' of ↵David S. Miller2009-09-02
|\ \ | | | | | | | | | | | | | | | | | | master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6 Conflicts: drivers/net/yellowfin.c
| * | pkt_sched: Revert tasklet_hrtimer changes.David S. Miller2009-09-01
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | These are full of unresolved problems, mainly that conversions don't work 1-1 from hrtimers to tasklet_hrtimers because unlike hrtimers tasklets can't be killed from softirq context. And when a qdisc gets reset, that's exactly what we need to do here. We'll work this out in the net-next-2.6 tree and if warranted we'll backport that work to -stable. This reverts the following 3 changesets: a2cb6a4dd470d7a64255a10b843b0d188416b78f ("pkt_sched: Fix bogon in tasklet_hrtimer changes.") 38acce2d7983632100a9ff3fd20295f6e34074a8 ("pkt_sched: Convert CBQ to tasklet_hrtimer.") ee5f9757ea17759e1ce5503bdae2b07e48e32af9 ("pkt_sched: Convert qdisc_watchdog to tasklet_hrtimer") Signed-off-by: David S. Miller <davem@davemloft.net>
| * | pkt_sched: Convert qdisc_watchdog to tasklet_hrtimerDavid S. Miller2009-08-22
| | | | | | | | | | | | | | | | | | | | | | | | None of this stuff should execute in hw IRQ context, therefore use a tasklet_hrtimer so that it runs in softirq context. Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Thomas Gleixner <tglx@linutronix.de>
| * | net: restore gnet_stats_basic to previous definitionEric Dumazet2009-08-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In 5e140dfc1fe87eae27846f193086724806b33c7d "net: reorder struct Qdisc for better SMP performance" the definition of struct gnet_stats_basic changed incompatibly, as copies of this struct are shipped to userland via netlink. Restoring old behavior is not welcome, for performance reason. Fix is to use a private structure for kernel, and teach gnet_stats_copy_basic() to convert from kernel to user land, using legacy structure (struct gnet_stats_basic) Based on a report and initial patch from Michael Spang. Reported-by: Michael Spang <mspang@csclub.uwaterloo.ca> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | net: make neigh_ops constantStephen Hemminger2009-09-01
| | | | | | | | | | | | | | | | | | | | | | | | These tables are never modified at runtime. Move to read-only section. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | RTO connection timeout: coding style fixes and commentsDamian Lukowski2009-09-01
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch affects the retransmits_timed_out() function. Changes: 1) Variables have more meaningful names 2) retransmits_timed_out() has an introductionary comment. 3) Small coding style changes. Signed-off-by: Damian Lukowski <damian@tvk.rwth-aachen.de> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | netns: embed ip6_dst_ops directlyAlexey Dobriyan2009-09-01
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | struct net::ipv6.ip6_dst_ops is separatedly dynamically allocated, but there is no fundamental reason for it. Embed it directly into struct netns_ipv6. For that: * move struct dst_ops into separate header to fix circular dependencies I honestly tried not to, it's pretty impossible to do other way * drop dynamical allocation, allocate together with netns For a change, remove struct dst_ops::dst_net, it's deducible by using container_of() given dst_ops pointer. Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | Revert Backoff [v3]: Calculate TCP's connection close threshold as a time value.Damian Lukowski2009-09-01
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | RFC 1122 specifies two threshold values R1 and R2 for connection timeouts, which may represent a number of allowed retransmissions or a timeout value. Currently linux uses sysctl_tcp_retries{1,2} to specify the thresholds in number of allowed retransmissions. For any desired threshold R2 (by means of time) one can specify tcp_retries2 (by means of number of retransmissions) such that TCP will not time out earlier than R2. This is the case, because the RTO schedule follows a fixed pattern, namely exponential backoff. However, the RTO behaviour is not predictable any more if RTO backoffs can be reverted, as it is the case in the draft "Make TCP more Robust to Long Connectivity Disruptions" (http://tools.ietf.org/html/draft-zimmermann-tcp-lcd). In the worst case TCP would time out a connection after 3.2 seconds, if the initial RTO equaled MIN_RTO and each backoff has been reverted. This patch introduces a function retransmits_timed_out(N), which calculates the timeout of a TCP connection, assuming an initial RTO of MIN_RTO and N unsuccessful, exponentially backed-off retransmissions. Whenever timeout decisions are made by comparing the retransmission counter to some value N, this function can be used, instead. The meaning of tcp_retries2 will be changed, as many more RTO retransmissions can occur than the value indicates. However, it yields a timeout which is similar to the one of an unpatched, exponentially backing off TCP in the same scenario. As no application could rely on an RTO greater than MIN_RTO, there should be no risk of a regression. Signed-off-by: Damian Lukowski <damian@tvk.rwth-aachen.de> Acked-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | Revert Backoff [v3]: Revert RTO on ICMP destination unreachableDamian Lukowski2009-09-01
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Here, an ICMP host/network unreachable message, whose payload fits to TCP's SND.UNA, is taken as an indication that the RTO retransmission has not been lost due to congestion, but because of a route failure somewhere along the path. With true congestion, a router won't trigger such a message and the patched TCP will operate as standard TCP. This patch reverts one RTO backoff, if an ICMP host/network unreachable message, whose payload fits to TCP's SND.UNA, arrives. Based on the new RTO, the retransmission timer is reset to reflect the remaining time, or - if the revert clocked out the timer - a retransmission is sent out immediately. Backoffs are only reverted, if TCP is in RTO loss recovery, i.e. if there have been retransmissions and reversible backoffs, already. Changes from v2: 1) Renaming of skb in tcp_v4_err() moved to another patch. 2) Reintroduced tcp_bound_rto() and __tcp_set_rto(). 3) Fixed code comments. Signed-off-by: Damian Lukowski <damian@tvk.rwth-aachen.de> Acked-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | dcbnl: Add support for setapp/getapp to netdev dcbnl_rtnl_opsYi Zou2009-09-01
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Adds support of dcbnl setapp/getapp to dcbnl_rtnl_ops in netdev to allow LLDs to implement their corresponding dcbnl setapp/getapp ops to support the IEEE 802.1Q DCBX setapp/getapp commands. Signed-off-by: Yi Zou <yi.zou@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>
* | | Merge branch 'master' of ↵David S. Miller2009-08-31
|\ \ \ | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/holtmann/bluetooth-next-2.6
| * | | Bluetooth: Handle L2CAP case when the remote receiver is busyGustavo F. Padovan2009-08-26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Implement all issues related to RemoteBusy in the RECV state table. Signed-off-by: Gustavo F. Padovan <gustavo@las.ic.unicamp.br> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| * | | Bluetooth: Fix rejected connection not disconnecting ACL linkLuiz Augusto von Dentz2009-08-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When using DEFER_SETUP on a RFCOMM socket, a SABM frame triggers authorization which when rejected send a DM response. This is fine according to the RFCOMM spec: the responding implementation may replace the "proper" response on the Multiplexer Control channel with a DM frame, sent on the referenced DLCI to indicate that the DLCI is not open, and that the responder would not grant a request to open it later either. But some stacks doesn't seems to cope with this leaving DLCI 0 open after receiving DM frame. To fix it properly a timer was introduced to rfcomm_session which is used to set a timeout when the last active DLC of a session is unlinked, this will give the remote stack some time to reply with a proper DISC frame on DLCI 0 avoiding both sides sending DISC to each other on stacks that follow the specification and taking care of those who don't by taking down DLCI 0. Signed-off-by: Luiz Augusto von Dentz <luiz.dentz@openbossa.org> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| * | | Bluetooth: Full support for receiving L2CAP SREJ framesGustavo F. Padovan2009-08-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Support for receiving of SREJ frames as specified by the state table. Signed-off-by: Gustavo F. Padovan <gustavo@las.ic.unicamp.br> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| * | | Bluetooth: Add support for L2CAP SREJ exceptionGustavo F. Padovan2009-08-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When L2CAP loses an I-frame we send a SREJ frame to the transmitter side requesting the lost packet. This patch implement all Recv I-frame events on SREJ_SENT state table except the ones that deal with SendRej (the REJ exception at receiver side is yet not implemented). Signed-off-by: Gustavo F. Padovan <gustavo@las.ic.unicamp.br> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| * | | Bluetooth: Add support for FCS option to L2CAPGustavo F. Padovan2009-08-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Implement CRC16 check for L2CAP packets. FCS is used by Streaming Mode and Enhanced Retransmission Mode and is a extra check for the packet content. Using CRC16 is the default, L2CAP won't use FCS only when both side send a "No FCS" request. Initially based on a patch from Nathan Holstein <nathan@lampreynetworks.com> Signed-off-by: Gustavo F. Padovan <gustavo@las.ic.unicamp.br> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| * | | Bluetooth: Add support for Retransmission and Monitor TimersGustavo F. Padovan2009-08-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | L2CAP uses retransmission and monitor timers to inquiry the other side about unacked I-frames. After sending each I-frame we (re)start the retransmission timer. If it expires, we start a monitor timer that send a S-frame with P bit set and wait for S-frame with F bit set. If monitor timer expires, try again, at a maximum of L2CAP_DEFAULT_MAX_TX. Signed-off-by: Gustavo F. Padovan <gustavo@las.ic.unicamp.br> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| * | | Bluetooth: Initial support for retransmission of packets with REJ framesGustavo F. Padovan2009-08-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When receiving an I-frame with unexpected txSeq, receiver side start the recovery procedure by sending a REJ S-frame to the transmitter side. So the transmitter can re-send the lost I-frame. This patch just adds a basic support for retransmission, it doesn't mean that ERTM now has full support for packet retransmission. Signed-off-by: Gustavo F. Padovan <gustavo@las.ic.unicamp.br> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| * | | Bluetooth: Add support for Segmentation and Reassembly of SDUsGustavo F. Padovan2009-08-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ERTM should use Segmentation and Reassembly to break down a SDU in many PDUs on sending data to the other side. On sending packets we queue all 'segments' until end of segmentation and just the add them to the queue for sending. On receiving we create a new SKB with the SDU reassembled. Initially based on a patch from Nathan Holstein <nathan@lampreynetworks.com> Signed-off-by: Gustavo F. Padovan <gustavo@las.ic.unicamp.br> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| * | | Bluetooth: Add initial support for ERTM packets transfersGustavo F. Padovan2009-08-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch adds support for ERTM transfers, without retransmission, with txWindow up to 63 and with acknowledgement of packets received. Now the packets are queued before call l2cap_do_send(), so packets couldn't be sent at the time we call l2cap_sock_sendmsg(). They will be sent in an asynchronous way on later calls of l2cap_ertm_send(). Besides if an error occurs on calling l2cap_do_send() we disconnect the channel. Initially based on a patch from Nathan Holstein <nathan@lampreynetworks.com> Signed-off-by: Gustavo F. Padovan <gustavo@las.ic.unicamp.br> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| * | | Bluetooth: Add configuration support for ERTM and Streaming modeGustavo F. Padovan2009-08-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add support to config_req and config_rsp to configure ERTM and Streaming mode. If the remote device specifies ERTM or Streaming mode, then the same mode is proposed. Otherwise ERTM or Basic mode is used. And in case of a state 2 device, the remote device should propose the same mode. If not, then the channel gets disconnected. Signed-off-by: Gustavo F. Padovan <gustavo@las.ic.unicamp.br> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| * | | Bluetooth: Allow setting of L2CAP ERTM via socket optionMarcel Holtmann2009-08-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | To enable Enhanced Retransmission mode it needs to be set via a socket option. A different mode can be set on a socket, but on listen() and connect() the mode is checked and ERTM is only allowed if it is enabled via the module parameter. Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| * | | Bluetooth: Convert hdev->req_lock to a mutexThomas Gleixner2009-08-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | hdev->req_lock is used as mutex so make it a mutex. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| * | | Bluetooth: Add extra device reference counting for connectionsMarcel Holtmann2009-08-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The device model itself has no real usable reference counting at the moment and this causes problems if parents are deleted before their children. The device model itself handles the memory details of this correctly, but the uevent order is not consistent. This causes various problems for systems like HAL or even X. So until device_put() does a proper cleanup, the device for Bluetooth connection will be protected with an extra reference counting to ensure the correct order of uevents when connections are terminated. This is not an automatic feature. Higher Bluetooth layers like HIDP or BNEP should grab this new reference to ensure that their uevents are send before the ones from the parent device. Based on a report by Brian Rogers <brian@xyzw.org> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
* | | | tcp: keepalive cleanupsEric Dumazet2009-08-29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Introduce keepalive_probes(tp) helper, and use it, like keepalive_time_when(tp) and keepalive_intvl_when(tp) Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | | mac80211: remove ieee80211_rx namespace hackJohn W. Linville2009-08-28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | With the libipw naming scheme change, it is no longer necessary for mac80211 to avoid the ieee80211_rx name clash. Reported-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* | | | libipw: switch from ieee80211_* to libipw_* naming policyJohn W. Linville2009-08-28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This eliminates the dual definition of ieee80211_channel (and possibly others), further clarifying who defines what and paving the way for inclusion of cfg80211.h. Signed-off-by: John W. Linville <linville@tuxdriver.com>
* | | | Merge branch 'for-next' of ↵David S. Miller2009-08-23
|\ \ \ \ | |/ / / |/| | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/lowpan/lowpan
| * | | ieee802154: add support for channel pages from IEEE 802.15.4-2006Dmitry Eremin-Solenikov2009-08-19
| | | | | | | | | | | | | | | | | | | | | | | | | | | | IEEE 802.15.4-2006 adds new concept: channel pages, which can contain several channels. Add support for channel pages in the API and in the fakehard driver. Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
| * | | ieee802154: add a sysfs representation of WPAN master devicesDmitry Eremin-Solenikov2009-08-19
| | | | | | | | | | | | | | | | | | | | | | | | Add a sysfs/in-kernel representation of LR-WPAN master devices. Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
* | | | cfg80211: fix dangling scan request checkingJohannes Berg2009-08-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | My patch "cfg80211: fix deadlock" broke the code it was supposed to fix, the scan request checking. But it's not trivial to put it back the way it was, since the original patch had a deadlock. Now do it in a completely new way: queue the check off to a work struct, where we can freely lock. But that has some more complications, like needing to wait for it to be done before the wiphy/rdev can be destroyed, so some code is required to handle that. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* | | | mac80211: remove deprecated APIJohannes Berg2009-08-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | All but two drivers have now stopped using the two deprecated members radio_enabled and beacon_int, so it's about time to remove them for good. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Acked-by: Kalle Valo <kalle.valo@iki.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* | | | mac80211: allow configure_filter callback to sleepJohannes Berg2009-08-20
|/ / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Over time, a whole bunch of drivers have come up with their own scheme to delay the configure_filter operation to a workqueue. To be able to simplify things, allow configure_filter to sleep, and add a new prepare_multicast callback that drivers that need the multicast address list implement. This new callback must be atomic, but most drivers either don't care or just calculate a hash which can be done atomically and then uploaded to the hardware non-atomically. A cursory look suggests that at76c50x-usb, ar9170, mwl8k (which is actually very broken now), rt2x00, wl1251, wl1271 and zd1211 should make use of this new capability. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* | | cfg80211: allow driver to override PS defaultJohannes Berg2009-08-14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Sometimes drivers might have a good reason to override the PS default, like iwlwifi right now where it affects RX performance significantly at this point. This will allow them to override the default, if desired, in a way that users can still change it according to their trade-off choices, not the driver's, like would happen if the driver just disabled PS completely then. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* | | mac80211: remove max_bandwidthPat Erley2009-08-14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This removes the max_bandwidth attribute. It is only ever written to, and is duplicated by max_bandwidth_khz in the regulatory code. Signed-off-by: Pat Erley <pat-lkml@erley.org> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* | | cfg80211: fix alignment problem in scan requestJohannes Berg2009-08-14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The memory layout for scan requests was rather wrong, we put the scan SSIDs before the channels which could lead to the channel pointers being unaligned in memory. It turns out that using a pointer to the channel array isn't necessary anyway since we can embed a zero-length array into the struct. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* | | mac80211: allow DMA optimisationJohannes Berg2009-08-14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If we have a lot of frames to transmit at once, for instance with fragmentation, it can be an optimisation to only tell the DMA engine about them on the last fragment/frame to avoid banging the IO too much. This patch allows implementation such an optimisation by telling the driver when more frames can be expected. Currently, this is used by mac80211 only on fragmented frames, but could also be used in the future on other frames when the queue was full and there are multiple frames pending. Note that drivers need to be careful when using this flag, they need to kick their DMA engines not just when this flag is clear, but also when the queue gets full so that progress can be made. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* | | mac80211: document TX powersave filter requirementsJohannes Berg2009-08-14
| | | | | | | | | | | | | | | | | | | | | | | | This documents what's required to implement that TX powersave filter properly wrt. handling hardware queues. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* | | mac80211: explain TX retry and statusJohannes Berg2009-08-14
| | | | | | | | | | | | | | | | | | | | | | | | Add some more documentation including an example so that it's clearer what should be done for TX retries. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* | | nl80211: add generation number to all dumpsJohannes Berg2009-08-14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In order for userspace to be able to figure out whether it obtained a consistent snapshot of data or not when using netlink dumps, we need to have a generation number in each dump message that indicates whether the list has changed or not -- its value is arbitrary. This patch adds such a number to all dumps, this needs some mac80211 involvement to keep track of a generation number to start with when adding/removing mesh paths or stations. The wiphy and netdev lists can be fully handled within cfg80211, of course, but generation numbers need to be stored there as well. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* | | cfg80211: use reassociation when possibleJohannes Berg2009-08-14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | With the move of everything related to the SME from mac80211 to cfg80211, we lost the ability to send reassociation frames. This adds them back, but only for wireless extensions. With the userspace SME, it shall control assoc vs. reassoc (it already can do so with the nl80211 interface). I haven't touched the connect() implementation, so it is not possible to reassociate with the nl80211 connect primitive. I think that should be done with the NL80211_CMD_ROAM command, but we'll have to see how that can be handled in the future, especially with fullmac chips. This patch addresses only the immediate regression we had in mac80211, which previously sent reassoc. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* | | nl802154: support START-CONFIRM primitiveDmitry Baryshkov2009-08-13
| | | | | | | | | | | | | | | Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | af_ieee802154: add support for WANT_ACK socket optionDmitry Baryshkov2009-08-13
| | | | | | | | | | | | | | | Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | af_ieee802154: drop IEEE802154_SIOC_ADD_SLAVE declarationDmitry Baryshkov2009-08-12
| | | | | | | | | | | | | | | | | | | | | | | | IEEE802154_SIOC_ADD_SLAVE was used to allocate 802.15.4 interfaces on the top of radio. It's not used anymore, drop it. Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
generated by cgit v1.2.2 (git 2.25.0) at 2025-03-21 11:51:35 -0400