summaryrefslogtreecommitdiff
path: root/include (follow)
Commit message (Collapse)AuthorAge
* rhashtable: Rip out obsolete out-of-line interfaceHerbert Xu2015-03-20
| | | | | | | | | Now that all rhashtable users have been converted over to the inline interface, this patch removes the unused out-of-line interface. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
* rhashtable: Allow hash/comparison functions to be inlinedHerbert Xu2015-03-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch deals with the complaint that we make indirect function calls on the fast paths unnecessarily in rhashtable. We resolve it by moving the fast paths into inline functions that take struct rhashtable_param (which obviously must be the same set of parameters supplied to rhashtable_init) as an argument. The only remaining indirect call is to obj_hashfn (or key_hashfn it obj_hashfn is unset) on the rehash as well as the insert-during- rehash slow path. This patch also extends the support of vairable-length keys to include those where the key is fixed but scattered in the object. For example, in netlink we want to key off the namespace and the portid but they're not next to each other. This patch does this by directly using the object hash function as the indicator of whether the key is accessible or not. It also adds a new function obj_cmpfn to compare a key against an object. This means that the caller no longer needs to supply explicit compare functions. All this is done in a backwards compatible manner so no existing users are affected until they convert to the new interface. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
* rhashtable: Make rhashtable_init params argument constHerbert Xu2015-03-20
| | | | | | | | | | | | | This patch marks the rhashtable_init params argument const as there is no reason to modify it since we will always make a copy of it in the rhashtable. This patch also fixes a bug where we don't actually round up the value of min_size unless it is less than HASH_MIN_SIZE. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Acked-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
* net: increase sk_[max_]ack_backlogEric Dumazet2015-03-20
| | | | | | | | | | | | | | | | | | | | | | sk_ack_backlog & sk_max_ack_backlog were 16bit fields, meaning listen() backlog was limited to 65535. It is time to increase the width to allow much bigger backlog, if admins change /proc/sys/net/core/somaxconn & /proc/sys/net/ipv4/tcp_max_syn_backlog default values. Tested: echo 5000000 >/proc/sys/net/core/somaxconn echo 5000000 >/proc/sys/net/ipv4/tcp_max_syn_backlog Ran a SYNFLOOD test against a listener using listen(fd, 5000000) myhost~# grep request_sock_TCP /proc/slabinfo request_sock_TCP 4185642 4411940 304 13 1 : tunables 54 27 8 : slabdata 339380 339380 0 Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* inet: get rid of central tcp/dccp listener timerEric Dumazet2015-03-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | One of the major issue for TCP is the SYNACK rtx handling, done by inet_csk_reqsk_queue_prune(), fired by the keepalive timer of a TCP_LISTEN socket. This function runs for awful long times, with socket lock held, meaning that other cpus needing this lock have to spin for hundred of ms. SYNACK are sent in huge bursts, likely to cause severe drops anyway. This model was OK 15 years ago when memory was very tight. We now can afford to have a timer per request sock. Timer invocations no longer need to lock the listener, and can be run from all cpus in parallel. With following patch increasing somaxconn width to 32 bits, I tested a listener with more than 4 million active request sockets, and a steady SYNFLOOD of ~200,000 SYN per second. Host was sending ~830,000 SYNACK per second. This is ~100 times more what we could achieve before this patch. Later, we will get rid of the listener hash and use ehash instead. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* inet: drop prev pointer handling in request sockEric Dumazet2015-03-20
| | | | | | | | | | | When request sock are put in ehash table, the whole notion of having a previous request to update dl_next is pointless. Also, following patch will get rid of big purge timer, so we want to delete a request sock without holding listener lock. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* Merge branch 'for-upstream' of ↵David S. Miller2015-03-19
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next Johan Hedberg says: ==================== pull request: bluetooth-next 2015-03-19 This wont the last 4.1 bluetooth-next pull request, but we've piled up enough patches in less than a week that I wanted to save you from a single huge "last-minute" pull somewhere closer to the merge window. The main changes are: - Simultaneous LE & BR/EDR discovery support for HW that can do it - Complete LE OOB pairing support - More fine-grained mgmt-command access control (normal user can now do harmless read-only operations). - Added RF power amplifier support in cc2520 ieee802154 driver - Some cleanups/fixes in ieee802154 code Please let me know if there are any issues pulling. Thanks. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
| * cc2520: Add support for CC2591 amplifier.Brad Campbell2015-03-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The TI CC2521 is an RF power amplifier that is designed to interface with the CC2520. Conveniently, it directly interfaces with the CC2520 and does not require any pins to be connected to a microcontroller/processor. Adding a CC2591 increases the CC2520's range, which is useful for border router and other wall-powered applications. Using the CC2591 with the CC2520 requires configuring the CC2520 GPIOs that are connected to the CC2591 to correctly set the CC2591 into TX and RX modes. Further, TI recommends that the CC2520_TXPOWER and CC2520_AGCCTRL1 registers are set differently to maximize the CC2591's performance. These settings are covered in TI Application Note AN065. This patch adds an optional `amplified` field to the cc2520 entry in the device tree. If present, the CC2520 will be configured to operate with a CC2591. The expected pin mapping is: CC2520 GPIO0 --> CC2591 EN CC2520 GPIO5 --> CC2591 PAEN Signed-off-by: Brad Campbell <bradjc5@gmail.com> Acked-by: Varka Bhadram <varkabhadram@gmail.com> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| * Bluetooth: Move generic mgmt command dispatcher to hci_sock.cJohan Hedberg2015-03-17
| | | | | | | | | | | | | | | | | | | | | | The mgmt.c file should be reserved purely for HCI_CHANNEL_CONTROL. The mgmt_control() function in it is already completely generic and has a single user in hci_sock.c. This patch moves the function there and renames it a bit more appropriately to hci_mgmt_cmd() (as it's a command dispatcher). Signed-off-by: Johan Hedberg <johan.hedberg@intel.com> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| * Bluetooth: Add hdev_init callback for HCI channelsJohan Hedberg2015-03-17
| | | | | | | | | | | | | | | | | | | | In order to make the mgmt command handling more generic we can't have a direct call to mgmt_init_hdev() from mgmt_control(). This patch adds a new callback to struct hci_mgmt_chan. And sets it to point to the mgmt_init_hdev() function for the HCI_CHANNEL_CONTROL instance. Signed-off-by: Johan Hedberg <johan.hedberg@intel.com> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| * Bluetooth: Add helper to get HCI channel of a socketJohan Hedberg2015-03-17
| | | | | | | | | | | | | | | | | | We'll need to have access to which HCI channel a socket is bound to, in order to manage pending mgmt commands in clean way. This patch adds a helper for the purpose. Signed-off-by: Johan Hedberg <johan.hedberg@intel.com> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| * Bluetooth: Introduce HCI_QUIRK_SIMULTANEOUS_DISCOVERYJakub Pawlowski2015-03-17
| | | | | | | | | | | | | | | | | | | | | | | | | | Some controllers allow both LE scan and BR/EDR inquiry to run at the same time, while others allow only one, LE SCAN or BR/EDR inquiry at given time. Since this is specific to each controller, add a new quirk setting that allows drivers to tell the core wether given controller can do both LE scan and BR/EDR inquiry at same time. Signed-off-by: Jakub Pawlowski <jpawlowski@google.com> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Add support for Local OOB Extended Data Update eventsMarcel Holtmann2015-03-17
| | | | | | | | | | | | | | | | | | | | When a different user requests a new set of local out-of-band data, then inform all previous users that the data has been updated. To limit the scope of users, the updates are limited to previous users. If a user has never requested out-of-band data, it will also not see the update. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Add constants for LE SC Confirmation and Random valuesMarcel Holtmann2015-03-16
| | | | | | | | | | | | | | | | | | The LE Secure Connections Confirmation Value and LE Secure Connections Random Value contants are required for the out-of-band data and so just define them. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Remove unneeded HCI_CONN_REMOTE_OOB connection flagMarcel Holtmann2015-03-16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | The HCI_CONN_REMOTE_OOB connection flag is used to indicate if the pairing initiator has provided out-of-band data. However since that value is no longer used in any decision making, just remove it. It is actually unclear what purpose the OOB data present field from the HCI IO Capability Response event serves in the first place. If either side provided out-of-band data, then that data will be used for pairing. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Add simple version of Read Local OOB Extended Data commandMarcel Holtmann2015-03-15
| | | | | | | | | | | | | | | | | | This adds support for the simplest possible version of Read Local OOB Extended Data management command. It includes all mandatory fields, but none of the actual pairing related ones. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Add defines for LE Bluetooth Device Address and LE RoleMarcel Holtmann2015-03-15
| | | | | | | | | | | | | | | | The OOB data requires to include LE Bluetooth Device Address and LE Role and so add the type constants for these fields. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Add simple version of Read Advertising Features commandMarcel Holtmann2015-03-15
| | | | | | | | | | | | | | | | This adds support for the simplest possible version of Read Advertising Features management command. It allows basic testing of the interface. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Use BIT(n) macro instead of manually encoding (1 << n)Marcel Holtmann2015-03-15
| | | | | | | | | | | | | | | | | | The flags for the management command table used manual encoding of bits in the form of (1 << n). It is however preferred to use BIT(n) macro instead. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Send global configuration updates to all management usersMarcel Holtmann2015-03-15
| | | | | | | | | | | | | | | | | | | | | | Changes to the global configuration updates like settings, class of device, name etc. can be received by every user. They are allowed to read them in the first place so provide the updates via events as well. Otherwise untrusted users start polling for updates and that is not a desired behavior. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Add support for trust verification of management commandsMarcel Holtmann2015-03-15
| | | | | | | | | | | | | | | | | | Check the required trust level of each management command with the trust level of the management socket. If it does not match up, then return the newly introduced permission denied error. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Add support for untrusted access to management commandsMarcel Holtmann2015-03-15
| | | | | | | | | | | | | | | | | | Some management commands are safe to be accessed from any user without special permissions. First step for allowing access to any of these commands from untrusted application is to mark them accordingly. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Add hci_sock_test_flag helper functionMarcel Holtmann2015-03-15
| | | | | | | | | | | | | | | | The management interface will need access to the socket flags and so provide a helper function for checking them. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Consolidate socket channel sending function back into oneMarcel Holtmann2015-03-15
| | | | | | | | | | | | | | | | | | With the introduction of trusted socket flag for control and monitor channels, it is now possible to use a single function for sending packets to these sockets. And with that consolidate the handling. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Introduce trusted flag for management control socketsMarcel Holtmann2015-03-15
| | | | | | | | | | | | | | | | | | Providing a global trusted flag for management control sockets provides an easy way for identifying sockets and imposing restriction on it. For now all management sockets are trusted since they require CAP_NET_ADMIN. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Add support for extended index management commandMarcel Holtmann2015-03-15
| | | | | | | | | | | | | | | | | | The Read Extended Contoller Index List command can be used for retrieving the complete list of local available controllers. This included configured, unconfigured and also AMP controllers. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Add support for extended index management eventsMarcel Holtmann2015-03-15
| | | | | | | | | | | | | | | | | | This introduces support for using Extended Index Added and Extended Index Removed events. These events contain the controller type and also the hardware bus information from the driver. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Use special function to send filter management index eventsMarcel Holtmann2015-03-15
| | | | | | | | | | | | | | | | | | | | For sending Index Added, Index Removed, Unconfigured Index Added and Unconfigured Index Removed managment events the new helper functions allows taking into account if these events are enabled for a certain management socket or not. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Provide hci_send_to_flagged_channel helper functionMarcel Holtmann2015-03-15
| | | | | | | | | | | | | | | | | | | | | | | | The hci_send_to_flagged_channel helper function can be used to send packets to all channels that have a certain HCI socket flag set. This is especially useful for managment events that are limited to sockets that have first enabled certain functionality. This allows for filtering of events without confusing existing users. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
| * Bluetooth: Add flags field and setting function for HCI socketsMarcel Holtmann2015-03-15
| | | | | | | | | | | | | | | | | | | | | | To filter out certain actions for certain HCI sockets introcuce a flags field that allows to configure specific settings on individual sockets. Since the hci_pinfo structure is private in hci_sock.c, provide helper functions for setting and clearing a given flag. Signed-off-by: Marcel Holtmann <marcel@holtmann.org> Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
* | bridge: add ageing_time, stp_state, priority over netlinkJörg Thalheim2015-03-18
| | | | | | | | | | Signed-off-by: Jörg Thalheim <joerg@higgsboson.tk> Signed-off-by: David S. Miller <davem@davemloft.net>
* | net: Fix high overhead of vlan sub-device teardown.David S. Miller2015-03-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When a networking device is taken down that has a non-trivial number of VLAN devices configured under it, we eat a full synchronize_net() for every such VLAN device. This is because of the call chain: NETDEV_DOWN notifier --> vlan_device_event() --> dev_change_flags() --> __dev_change_flags() --> __dev_close() --> __dev_close_many() --> dev_deactivate_many() --> synchronize_net() This is kind of rediculous because we already have infrastructure for batching doing operation X to a list of net devices so that we only incur one sync. So make use of that by exporting dev_close_many() and adjusting it's interfaace so that the caller can fully manage the batch list. Use this in vlan_device_event() and all the overhead goes away. Reported-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | net: add support for phys_port_nameDavid Ahern2015-03-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Similar to port id allow netdevices to specify port names and export the name via sysfs. Drivers can implement the netdevice operation to assist udev in having sane default names for the devices using the rule: $ cat /etc/udev/rules.d/80-net-setup-link.rules SUBSYSTEM=="net", ACTION=="add", ATTR{phys_port_name}!="", NAME="$attr{phys_port_name}" Use of phys_name versus phys_id was suggested-by Jiri Pirko. Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Jiri Pirko <jiri@resnulli.us> Acked-by: Scott Feldman <sfeldma@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | ipv4, ipv6: kill ip_mc_{join, leave}_group and ipv6_sock_mc_{join, drop}Marcelo Ricardo Leitner2015-03-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | in favor of their inner __ ones, which doesn't grab rtnl. As these functions need to operate on a locked socket, we can't be grabbing rtnl by then. It's too late and doing so causes reversed locking. So this patch: - move rtnl handling to callers instead while already fixing some reversed locking situations, like on vxlan and ipvs code. - renames __ ones to not have the __ mark: __ip_mc_{join,leave}_group -> ip_mc_{join,leave}_group __ipv6_sock_mc_{join,drop} -> ipv6_sock_mc_{join,drop} Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* | inet: request sock should init IPv6/IPv4 addressesEric Dumazet2015-03-18
| | | | | | | | | | | | | | | | In order to be able to use sk_ehashfn() for request socks, we need to initialize their IPv6/IPv4 addresses. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | inet: get rid of last __inet_hash_connect() argumentEric Dumazet2015-03-18
| | | | | | | | | | | | | | | | We now always call __inet_hash_nolisten(), no need to pass it as an argument. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | ipv6: get rid of __inet6_hash()Eric Dumazet2015-03-18
| | | | | | | | | | | | | | | | We can now use inet_hash() and __inet_hash() instead of private functions. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | inet: add IPv6 support to sk_ehashfn()Eric Dumazet2015-03-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Intent is to converge IPv4 & IPv6 inet_hash functions to factorize code. IPv4 sockets initialize sk_rcv_saddr and sk_v6_daddr in this patch, thanks to new sk_daddr_set() and sk_rcv_saddr_set() helpers. __inet6_hash can now use sk_ehashfn() instead of a private inet6_sk_ehashfn() and will simply use __inet_hash() in a following patch. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | net: introduce sk_ehashfn() helperEric Dumazet2015-03-18
| | | | | | | | | | | | | | | | | | | | Goal is to unify IPv4/IPv6 inet_hash handling, and use common helpers for all kind of sockets (full sockets, timewait and request sockets) inet_sk_ehashfn() becomes sk_ehashfn() but still only copes with IPv4 Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | netns: constify net_hash_mix() and various callersEric Dumazet2015-03-18
| | | | | | | | | | | | | | | | const qualifiers ease code review by making clear which objects are not written in a function. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | net/mlx4_core: Add basic support for QP max-rate limitingOr Gerlitz2015-03-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add the low-level device commands and definitions used for QP max-rate limiting. This is done through the following elements: - read rate-limit device caps in QUERY_DEV_CAP: number of different rates and the min/max rates in Kbs/Mbs/Gbs units - enhance the QP context struct to contain rate limit units and value - allow to do run time rate-limit setting to QPs through the update-qp firmware command - QP rate-limiting is disallowed for VFs Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | net: Add max rate tx queue attributeJohn Fastabend2015-03-18
| | | | | | | | | | | | | | | | | | | | | | | | | | This adds a tx_maxrate attribute to the tx queue sysfs entry allowing for max-rate limiting. Along with DCB-ETS and BQL this provides another knob to tune queue performance. The limit units are Mbps. By default it is disabled. To disable the rate limitation after it has been set for a queue, it should be set to zero. Signed-off-by: John Fastabend <john.r.fastabend@intel.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | rhashtable: Remove max_shift and min_shiftHerbert Xu2015-03-18
| | | | | | | | | | | | | | | | Now that nobody uses max_shift and min_shift, we can safely remove them. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
* | rhashtable: Introduce max_size/min_sizeHerbert Xu2015-03-18
| | | | | | | | | | | | | | | | This patch adds the parameters max_size and min_size which are meant to replace max_shift and min_shift. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
* | rhashtable: Remove shift from bucket_tableHerbert Xu2015-03-18
| | | | | | | | | | | | | | Keeping both size and shift is silly. We only need one. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
* | inet: fix request sock refcountingEric Dumazet2015-03-17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | While testing last patch series, I found req sock refcounting was wrong. We must set skc_refcnt to 1 for all request socks added in hashes, but also on request sockets created by FastOpen or syncookies. It is tricky because we need to defer this initialization so that future RCU lookups do not try to take a refcount on a not yet fully initialized request socket. Also get rid of ireq_refcnt alias. Signed-off-by: Eric Dumazet <edumazet@google.com> Fixes: 13854e5a6046 ("inet: add proper refcounting to request sock") Signed-off-by: David S. Miller <davem@davemloft.net>
* | tcp: rename struct tcp_request_sock listenerEric Dumazet2015-03-17
| | | | | | | | | | | | | | | | | | The listener field in struct tcp_request_sock is a pointer back to the listener. We now have req->rsk_listener, so TCP only needs one boolean and not a full pointer. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | inet: add rsk_listener field to struct request_sockEric Dumazet2015-03-17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Once we'll be able to lookup request sockets in ehash table, we'll need to get access to listener which created this request. This avoid doing a lookup to find the listener, which benefits for a more solid SO_REUSEPORT, and is needed once we no longer queue request sock into a listener private queue. Note that 'struct tcp_request_sock'->listener could be reduced to a single bit, as TFO listener should match req->rsk_listener. TFO will no longer need to hold a reference on the listener. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | inet: uninline inet_reqsk_alloc()Eric Dumazet2015-03-17
| | | | | | | | | | | | | | inet_reqsk_alloc() is becoming fat and should not be inlined. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | inet: add sk_listener argument to inet_reqsk_alloc()Eric Dumazet2015-03-17
| | | | | | | | | | | | | | | | | | | | | | listener socket can be used to set net pointer, and will be later used to hold a reference on listener. Add a const qualifier to first argument (struct request_sock_ops *), and factorize all write_pnet(&ireq->ireq_net, sock_net(sk)); Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
generated by cgit v1.2.3 (git 2.25.1) at 2026-04-24 18:32:33 +0000