summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* mac80211: track assigned vifs in chanctxMichal Kazior2014-04-253-0/+9
| | | | | | | | | | | | This can be useful. Provides a more straghtforward way to iterate over interfaces bound to a given chanctx and allows tracking chanctx usage explicitly. The structure is protected by local->chanctx_mtx. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: add support for radar detection for reservationsMichal Kazior2014-04-252-2/+7
| | | | | | | | | | Initial chanctx reservation code wasn't aware of radar detection requirements. This is necessary for chanctx reservations to be used for channel switching in the future. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: prevent chanctx overcommitMichal Kazior2014-04-251-1/+23
| | | | | | | | | | | | | Do not allocate more channel contexts than a driver is capable for currently matching interface combination. This allows the ieee80211_vif_reserve_chanctx() to act as a guard against breaking interface combinations. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: add max channel calculation utility functionMichal Kazior2014-04-252-0/+43
| | | | | | | | | The utility function has no uses yet. It is aimed at future chanctx reservation management and channel switching. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: allow drivers to iterate over matching combinationsMichal Kazior2014-04-252-7/+64
| | | | | | | | | | | | | | The patch splits cfg80211_check_combinations() into an iterator function and a simple iteration user. This makes it possible for drivers to asses how many channels can use given iftype setup. This in turn can be used for future multi-interface/multi-channel channel switching. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: Fix GO Concurrent relaxation on UNII-3Ilan Peer2014-04-251-1/+17
| | | | | | | | | | | | | | | | | At some locations, channels 149-165 are considered a single bundle, while at some other locations, e.g., Indonesia, channels 149-161 are considered a single bundle, while channel 165 belongs to a different bundle. This means that: 1. A station interface connection to an AP on channel 165 allows the instantiation of a P2P GO on channels 149-165. 2. A station interface connection to an AP on channels 149-161 does NOT allow the instantiation of a P2P GO on channel 165. Fix this. Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: mesh: always use the latest target_snBob Copeland2014-04-221-1/+2
| | | | | | | | | | | | When a path target responds to a path request, its response always contains the most up-to-date information; accordingly, it should use the latest target_sn, regardless of net_traversal_jiffies(). Otherwise, only the first path response is considered when constructing a path, as it will have the highest target_sn of all replies during that period. Signed-off-by: Bob Copeland <bob@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: fix mesh_add_rsn_ie IE finding loopBob Copeland2014-04-221-13/+8
| | | | | | | | | | | | Previously, the code to copy the RSN IE from the mesh config would increment its pointer by one in the loop instead of by the element length, so there was the potential for mistaking another IE's data fields as the RSN IE. cfg80211_find_ie() exists, so just use that. Signed-off-by: Bob Copeland <me@bobcopeland.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: mesh: use u16 return type for u16 getterBob Copeland2014-04-221-1/+1
| | | | | | | | | u16_field_get() is a simple wrapper around get_unaligned_le16(), and it is being assigned to a u16, so there's no need to promote to u32 in the middle. Signed-off-by: Bob Copeland <me@bobcopeland.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: Allow HT capa override to add 40 MHz intolerantJouni Malinen2014-04-222-2/+19
| | | | | | | | This can be useful for testing purposes to confirm valid AP behavior on HT 20/40 co-existence functionality. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: fix processing world regdomain when non modularLuis R. Rodriguez2014-04-221-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This allows processing of the last regulatory request when we determine its still pending. Without this if a regulatory request failed to get processed by userspace we wouldn't be able to re-process it later. An example situation that can lead to an unprocessed last_request is enabling cfg80211 to be built-in to the kernel, not enabling CFG80211_INTERNAL_REGDB and the CRDA binary not being available at the time the udev rule that kicks of CRDA triggers. In such a situation we want to let some cfg80211 triggers eventually kick CRDA for us again. Without this if the first cycle attempt to kick off CRDA failed we'd be stuck without the ability to change process any further regulatory domains. cfg80211 will trigger re-processing of the regulatory queue whenever schedule_work(&reg_work) is called, currently this happens when: * suspend / resume * disconnect * a beacon hint gets triggered (non DFS 5 GHz AP found) * a regulatory request gets added to the queue We don't have any specific opportunistic late boot triggers to address a late mount of where CRDA resides though, adding that should be done separately through another patch. Without an opportunistic fix then this fix relies at least one of the triggeres above to happen. Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Signed-off-by: Luis R. Rodriguez <mcgrof@suse.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: avoid freeing last_request while in flightArik Nemtsov2014-04-221-8/+16
| | | | | | | | | | | | | Avoid freeing the last request while it is being processed. This can happen in some cases if reg_work is kicked for some reason while the currently pending request is in flight. Cc: Sander Eikelenboom <linux@eikelenboom.it> Tested-by: Eliad Peller <eliad@wizery.com> Tested-by: Colleen Twitty <colleen@cozybit.com> Signed-off-by: Arik Nemtsov <arik@wizery.com> Signed-off-by: Luis R. Rodriguez <mcgrof@suse.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: reg: set DFS CAC time in case of custom regdJanusz Dziedzic2014-04-111-0/+7
| | | | | | | | | | | Set DFS CAC time also in case of using custom and strict regulatory from drivers. In other case we could have unset DFS CAC time directly after driver loaded and before issue regulatory set from user mode. Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: ignore invalid BSSIDs when looking for BSSesJohannes Berg2014-04-101-0/+2
| | | | | | | | | | When looking for a BSS matching given parameters, ignore invalid BSSIDs. This avoids, for example, trying to join an IBSS that has a multicast BSSID, which isn't supported by all drivers nor is it a valid configuration of the IBSS so better create a new one with a correctly chosen random BSSID. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: reject invalid IBSS BSSIDs in wext compat codeJohannes Berg2014-04-101-0/+3
| | | | | | | Don't allow using a multicast address as the BSSID, that isn't a valid configuration. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: remove unused wiphy argument from cfg80211_wext_freq()Zhao, Gang2014-04-105-8/+12
| | | | | | | | | | cfg80211_wext_freq() is declared in wext-compat.h, but its parameter struct wiphy's declaration is not included there. As the parameter isn't used, just remove it. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> [remove parameter instead of changing to netdev] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: fix some missing includesZhao, Gang2014-04-093-0/+5
| | | | | | | | Some header files in mac80211 don't include all the header files they require, fix that. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: Update conf_is_ht() to work properly with 5/10MHz channelsRostislav Lisovy2014-04-091-1/+3
| | | | | | | | | The channels with 5/10MHz bandwidth are not HT. We have to reflect this in conf_is_ht() function which returns whether the particular channel is HT or not. Signed-off-by: Rostislav Lisovy <rostislav.lisovy@fel.cvut.cz> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: update comment about WIPHY_FLAG_CUSTOM_REGULATORYKalle Valo2014-04-091-1/+1
| | | | | | | | | Commit a2f73b6c5db3c ("cfg80211: move regulatory flags to their own variable") renamed WIPHY_FLAG_CUSTOM_REGULATORY to REGULATORY_CUSTOM_REG, but missed to update one comment. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: Allow disabling LDPCPawel Kulakowski2014-04-092-1/+6
| | | | | | | This allows user-space (wpa_supplicant) to disable LDPC coding. Signed-off-by: Pawel Kulakowski <pawel.kulakowski@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: fix radar_detect combination checkingMichal Kazior2014-04-091-1/+1
| | | | | | | | | | All bits from radar_detect must match combination radar bitmask. Otherwise it is theoretically possible to lead into an invalid combination provided a driver reports strange combinations. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: allow reservation of a running chanctxLuciano Coelho2014-04-092-18/+68
| | | | | | | | | | | | | | | | | | | With single-channel drivers, we need to be able to change a running chanctx if we want to use chanctx reservation. Not all drivers may be able to do this, so add a flag that indicates support for it. Changing a running chanctx can also be used as an optimization in multi-channel drivers when the context needs to be reserved for future usage. Introduce IEEE80211_CHANCTX_RESERVED chanctx mode to mark a channel as reserved so nobody else can use it (since we know it's going to change). In the future, we may allow several vifs to use the same reservation as long as they plan to use the chanctx on the same future channel. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: implement chanctx reservationLuciano Coelho2014-04-092-34/+176
| | | | | | | | | | | | | | | | | | | | In order to support channel switch with multiple vifs and multiple contexts, we implement a concept of channel context reservation. This allows us to reserve a channel context to be used later. The reservation functionality is not tied directly to channel switch and may be used in other situations (eg. reserving a channel context during IBSS join). We first check if an existing compatible context exists and if it does, we reserve it. If there is no compatible context we create a new one and reserve it. Additionally, split ieee80211_vif_copy_chanctx_to_vlans() so we can call it while already holding the chanctx mutex. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: refactor ieee80211_assign/unassign_vif_chanctx into oneLuciano Coelho2014-04-091-42/+43
| | | | | | | | | | | Combine the functions into one, so that we can switch from one context to the other without having to unassign and assign separately. This is needed by the channel reservation functionality because otherwise we have a small period of time when the chanctx is set to NULL, which can cause problems if someone else is trying to dereference it. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: split ieee80211_vif_change_channel in twoLuciano Coelho2014-04-091-25/+33
| | | | | | | | | | ieee80211_vif_change_channel() locks chanctx_mtx. When implementing channel reservation for CS, we will need to call the function to change channel when the lock is already held, so split the part that requires the lock out and leave the locking in the original function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: protect AP VLAN list with local->mtxMichal Kazior2014-04-094-6/+14
| | | | | | | | | | | | | | | | | | | It was impossible to change chanctx of master AP for AP VLANs because the copy function requires RTNL which can't be simply taken in mac80211 code due to possible deadlocks. This is required for future chanctx reservation that re-bind vifs to new chanctx. This requires safe AP VLAN iteration without RTNL. Now VLANs can be iterated while holding either RTNL or local->mtx because the list is modified while holding both of these locks. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211/mac80211: move more combination checks to mac80211Luciano Coelho2014-04-097-42/+26
| | | | | | | | | | | | | | | Get rid of the cfg80211_can_add_interface() and cfg80211_can_change_interface() functions by moving that functionality to mac80211. With this patch all interface combination checks are now out of cfg80211 (except for the channel switch case which will be addressed in a future commit). Additionally, modify the ieee80211_check_combinations() function so that an undefined chandef can be passed, in order to use it before a channel is defined. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211/mac80211: move combination check to mac80211 for ibssLuciano Coelho2014-04-092-31/+29
| | | | | | | | Now that mac80211 can check the interface combinations itself, move the combinations check from cfg80211 to mac80211 when joining an IBSS. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211/mac80211: move interface counting for combination check to mac80211Luciano Coelho2014-04-0911-80/+110
| | | | | | | | | | | Move the counting part of the interface combination check from cfg80211 to mac80211. This is needed to simplify locking when the driver has to perform a combination check by itself (eg. with channel-switch). Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211/mac80211: refactor cfg80211_chandef_dfs_required()Luciano Coelho2014-04-096-62/+104
| | | | | | | | | Some interface types don't require DFS (such as STATION, P2P_CLIENT etc). In order to centralize these decisions, make cfg80211_chandef_dfs_required() take the iftype into consideration. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: refactor cfg80211_can_use_iftype_chan()Luciano Coelho2014-04-093-56/+96
| | | | | | | | | | | | | Separate the code that counts the interface types and channels from the code that check the interface combinations. The new function that checks for combinations is exported so it can be called by the drivers. This is done in preparation for moving the interface combinations checks out of cfg80211. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: Enable GO operation on indoor channelsIlan Peer2014-04-093-2/+20
| | | | | | | | | | Allow GO operation on a channel marked with IEEE80211_CHAN_INDOOR_ONLY iff there is a user hint indicating that the platform is operating in an indoor environment, i.e., the platform is a printer or media center device. Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: Add an option to hint indoor operationIlan Peer2014-04-094-12/+68
| | | | | | | | Add the option to hint the wireless core that it is operating in an indoor environment. Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: Enable GO operation on additional channelsIlan Peer2014-04-0910-16/+169
| | | | | | | | | | | | | | | | | | | | | Allow GO operation on a channel marked with IEEE80211_CHAN_GO_CONCURRENT iff there is an active station interface that is associated to an AP operating on the same channel in the 2 GHz band or the same UNII band (in the 5 GHz band). This relaxation is not allowed if the channel is marked with IEEE80211_CHAN_RADAR. Note that this is a permissive approach to the FCC definitions, that require a clear assessment that the device operating the AP is an authorized master, i.e., with radar detection and DFS capabilities. It is assumed that such restrictions are enforced by user space. Furthermore, it is assumed, that if the conditions that allowed for the operation of the GO on such a channel change, i.e., the station interface disconnected from the AP, it is the responsibility of user space to evacuate the GO from the channel. Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: Add Kconfig option for cellular BS hintsIlan Peer2014-04-092-1/+14
| | | | | | | | | Move the regulatory cellular base station hints support under a specific configuration option and make the option depend on CFG80211_CERTIFICATION_ONUS. Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: Add indoor only and GO concurrent channel attributesDavid Spinadel2014-04-094-0/+36
| | | | | | | | | | | | | | | | | | | | | | | | The FCC are clarifying some soft configuration requirements, which among other include the following: 1. Indoor operation, where a device can use channels requiring indoor operation, subject to that it can guarantee indoor operation, i.e., the device is connected to AC Power or the device is under the control of a local master that is acting as an AP and is connected to AC Power. 2. Concurrent GO operation, where devices may instantiate a P2P GO while they are under the guidance of an authorized master. For example, on a channel on which a BSS is connected to an authorized master, i.e., with DFS and radar detection capability in the UNII band. See https://apps.fcc.gov/eas/comments/GetPublishedDocument.html?id=327&tn=528122 Add support for advertising Indoor-only and GO-Concurrent channel properties. Signed-off-by: David Spinadel <david.spinadel@intel.com> Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: slightly clean up of cfg80211_sme_connect()Zhao, Gang2014-04-091-2/+1
| | | | | | | | | | | | | | Wdev->ssid_len has already been set in cfg80211_connect() and is equal to connect->ssid_len. Use wdev->ssid_len instead of connect->ssid_len so it will be consistent with previous ssid assignment statement. If bss is found in cfg80211_get_conn_bss(), wdev->conn->state is set to CFG80211_CONN_AUTHENTICATE_NEXT in there. So it's not needed to set it manually to CFG80211_CONN_AUTHENTICATE_NEXT if bss is found in that function. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: use RCU_INIT_POINTERMonam Agarwal2014-04-098-17/+17
| | | | | | | | | | | rcu_assign_pointer() ensures that the initialization of a structure is carried out before storing a pointer to that structure. However, in the case that NULL is assigned there's no structure to initialize so using RCU_INIT_POINTER instead is safe and more efficient. Signed-off-by: Monam Agarwal <monamagarwal123@gmail.com> [squash eight tiny patches, rewrite commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: regulatory: use RCU_INIT_POINTERMonam Agarwal2014-04-091-1/+1
| | | | | | | | | | | rcu_assign_pointer() ensures that the initialization of a structure is carried out before storing a pointer to that structure. However, in the case that NULL is assigned there's no structure to initialize so using RCU_INIT_POINTER instead is safe and more efficient. Signed-off-by: Monam Agarwal <monamagarwal123@gmail.com> [rewrite commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: add vif to flush callEmmanuel Grumbach2014-04-0920-30/+48
| | | | | | | | | | | This will allow the low level driver to make decision based on the vif such as queues etc... Since the vif might be NULL, we can't add it to the tracing functions. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> [fix staging rtl8821ae driver] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: allow userspace to take ownership of interfacesJohannes Berg2014-04-095-1/+91
| | | | | | | | | | | | | | | When dynamically creating interfaces from userspace, e.g. for P2P usage, such interfaces are usually owned by the process that created them, i.e. wpa_supplicant. Should wpa_supplicant crash, such interfaces will often cease operating properly and cause problems on restarting the process. To avoid this problem, introduce an ownership concept for interfaces. If an interface is owned by a netlink socket, then it will be destroyed if the netlink socket is closed for any reason, including if the process it belongs to crashed. This gives us a race-free way to get rid of any such interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: remove VLAIS usage from mac80211Jan-Simon Möller2014-04-091-18/+19
| | | | | | | | | | | | | | | | | | | | Replaced the use of a Variable Length Array In Struct (VLAIS) with a C99 compliant equivalent. This is the original VLAIS struct. struct { struct aead_request req; u8 priv[crypto_aead_reqsize(tfm)]; } aead_req; This patch instead allocates the appropriate amount of memory using a char array. The new code can be compiled with both gcc and clang. Signed-off-by: Jan-Simon Möller <dl9pf@gmx.de> Signed-off-by: Behan Webster <behanw@converseincode.com> [small style cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-nextLinus Torvalds2014-04-021387-30582/+99667
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pull networking updates from David Miller: "Here is my initial pull request for the networking subsystem during this merge window: 1) Support for ESN in AH (RFC 4302) from Fan Du. 2) Add full kernel doc for ethtool command structures, from Ben Hutchings. 3) Add BCM7xxx PHY driver, from Florian Fainelli. 4) Export computed TCP rate information in netlink socket dumps, from Eric Dumazet. 5) Allow IPSEC SA to be dumped partially using a filter, from Nicolas Dichtel. 6) Convert many drivers to pci_enable_msix_range(), from Alexander Gordeev. 7) Record SKB timestamps more efficiently, from Eric Dumazet. 8) Switch to microsecond resolution for TCP round trip times, also from Eric Dumazet. 9) Clean up and fix 6lowpan fragmentation handling by making use of the existing inet_frag api for it's implementation. 10) Add TX grant mapping to xen-netback driver, from Zoltan Kiss. 11) Auto size SKB lengths when composing netlink messages based upon past message sizes used, from Eric Dumazet. 12) qdisc dumps can take a long time, add a cond_resched(), From Eric Dumazet. 13) Sanitize netpoll core and drivers wrt. SKB handling semantics. Get rid of never-used-in-tree netpoll RX handling. From Eric W Biederman. 14) Support inter-address-family and namespace changing in VTI tunnel driver(s). From Steffen Klassert. 15) Add Altera TSE driver, from Vince Bridgers. 16) Optimizing csum_replace2() so that it doesn't adjust the checksum by checksumming the entire header, from Eric Dumazet. 17) Expand BPF internal implementation for faster interpreting, more direct translations into JIT'd code, and much cleaner uses of BPF filtering in non-socket ocntexts. From Daniel Borkmann and Alexei Starovoitov" * git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1976 commits) netpoll: Use skb_irq_freeable to make zap_completion_queue safe. net: Add a test to see if a skb is freeable in irq context qlcnic: Fix build failure due to undefined reference to `vxlan_get_rx_port' net: ptp: move PTP classifier in its own file net: sxgbe: make "core_ops" static net: sxgbe: fix logical vs bitwise operation net: sxgbe: sxgbe_mdio_register() frees the bus Call efx_set_channels() before efx->type->dimension_resources() xen-netback: disable rogue vif in kthread context net/mlx4: Set proper build dependancy with vxlan be2net: fix build dependency on VxLAN mac802154: make csma/cca parameters per-wpan mac802154: allow only one WPAN to be up at any given time net: filter: minor: fix kdoc in __sk_run_filter netlink: don't compare the nul-termination in nla_strcmp can: c_can: Avoid led toggling for every packet. can: c_can: Simplify TX interrupt cleanup can: c_can: Store dlc private can: c_can: Reduce register access can: c_can: Make the code readable ...
| * netpoll: Use skb_irq_freeable to make zap_completion_queue safe.Eric W. Biederman2014-04-011-1/+1
| | | | | | | | | | | | | | | | | | | | Replace the test in zap_completion_queue to test when it is safe to free skbs in hard irq context with skb_irq_freeable ensuring we only free skbs when it is safe, and removing the possibility of subtle problems. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * net: Add a test to see if a skb is freeable in irq contextEric W. Biederman2014-04-011-0/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently netpoll and skb_release_head_state assume that a skb is freeable in hard irq context except when skb->destructor is set. The reality is far from this. So add a function skb_irq_freeable to compute the full test and in the process be the living documentation of what the requirements are of actually freeing a skb in hard irq context. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * Merge tag 'linux-can-fixes-for-3.15-20140401' of ↵David S. Miller2014-04-015-162/+266
| |\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://gitorious.org/linux-can/linux-can linux-can-fixes-for-3.15-20140401 Marc Kleine-Budde says: ==================== this is a pull request of 16 patches for the 3.15 release cycle. Bjorn Van Tilt contributes a patch which fixes a memory leak in usb_8dev's usb_8dev_start_xmit()s error path. A patch by Robert Schwebel fixes a typo in the can documentation. The remaining patches all target the c_can driver. Two of them are by me; they add a missing netif_napi_del() and return value checking. Thomas Gleixner contributes 12 patches, which address several shortcomings in the driver like hardware initialisation, concurrency, message ordering and poor performance. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
| | * can: c_can: Avoid led toggling for every packet.Thomas Gleixner2014-04-011-3/+4
| | | | | | | | | | | | | | | | | | | | | | | | There is no point to toggle the RX led for every packet. Especially if we have a full FIFO we want to avoid everything we can. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
| | * can: c_can: Simplify TX interrupt cleanupThomas Gleixner2014-04-011-20/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The function loads the message object from the hardware to get the payload length. The previous patch stores that information in an array, so we can avoid the hardware access. Remove the hardware access and move the led toggle outside of the spinlocked region. Toggle the led only once when at least one packet has been received. Binary size shrinks along with the code Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
| | * can: c_can: Store dlc privateThomas Gleixner2014-04-012-27/+29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We can avoid the HW access in TX cleanup path for retrieving the DLC of the sent package if we store the DLC in a private array. Ideally this should be handled in the can_echo_skb functions, but I leave that exercise to the CAN folks. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
| | * can: c_can: Reduce register accessThomas Gleixner2014-04-011-34/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | commit 4ce78a838c (can: c_can: Speed up rx_poll function) hyped a performance improvement by reducing the access to the interrupt pending register from a dual 16 bit to a single 16 bit access. Wow! Thereby it crippled the driver to cast the 16 msg objects in stone, which is completly braindead as contemporary hardware has up to 128 message objects. Supporting larger object buffers is a major surgery, but it'd be definitely worth it especially as the driver does not support HW message filtering .... The logic of the "FIFO" implementation is to split the FIFO in half. For the lower half we read the buffers and clear the interrupt pending bit, but keep the newdat bit set, so the HW will queue above those buffers. When we read out the last low buffer then we reenable all the low half buffers by clearing the newdat bit. The upper half buffers clear the newdat and the interrupt pending bit right away as we know that the lower half bits are clear and give us a headstart against the hardware. Now the implementation is: transfer_message_object() read_object_and_put_into_skb(); if (obj < END_OF_LOW_BUF) clear_intpending(obj) else if (obj > END_OF_LOW_BUF) clear_intpending_and_newdat(obj) else if (obj == END_OF_LOW_BUF) clear_newdat_of_all_low_objects() The hardware allows to avoid most of the mess simply because we can tell the transfer_message_object() function to clear bits right away. So we can be clever and do: if (obj <= END_OF_LOW_BUF) ctrl = TRANSFER_MSG | CLEAR_INTPND; else ctrl = TRANSFER_MSG | CLEAR_INTPND | CLEAR_NEWDAT; transfer_message_object(ctrl) read_object_and_put_into_skb(); if (obj == END_OF_LOW_BUF) clear_newdat_of_all_low_objects() So we save a complete control operation on all message objects except the one which is the end of the low buffer. That's a few micro seconds per object. I'm not adding a boasting profile to that, simply because it's self explaining. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> [mkl: adjusted subject and commit message] Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-27 16:21:45 +0000