- 19 May, 2015 24 commits
-
-
Alexander Aring authored
This patch renames fakelb_dev_priv to fakelb_phy. We don't faking devices here, we fake wpan phys. This avoids also confusing with the variable priv, which is used several times in this driver to represent this structure. Signed-off-by:
Alexander Aring <alex.aring@gmail.com> Signed-off-by:
Marcel Holtmann <marcel@holtmann.org>
-
Alexander Aring authored
Iterate and removing items from a list, we should use list_for_each_entry_safe instead list_for_each_entry to avoid accidents by removing. Signed-off-by:
-
Alexander Aring authored
This patch change the default virtual wpan phys of fakelb driver from one to two. To have one virtual phy makes no sense, because it need at least two virtual phy's to speak to each other. Signed-off-by:
-
Alexander Aring authored
This patch adds support for the atusb transceiver. The current driver supports basic functionality only. Possible further tasks would be to sync functionality with the at86rf230 driver, because the atusb use internally an at86rf231 transceiver. Some of these features need a firmware update like AACK and ARET handling. I did small changes to this driver to work with xmit_async callback and setting of a random extended perm address. Signed-off-by:
-
Alexander Aring authored
Replace consume_skb with dev_consume_skb_any in ieee802154_xmit_complete which can be called in hard irq and other contexts. Signed-off-by:
-
Alexander Aring authored
This patch fix the complete callback for going into aret on. Currently we starting after a state change into aret on again a state change into aret on which is not necessary. Signed-off-by:
-
Martin Townsend authored
fakelb_hw_deliver creates a copy of the skb's header which can potentially return NULL so we now check for this before actually delivering to the 802.15.4 MAC layer. Signed-off-by:
Martin Townsend <martin.townsend@xsilon.com> Signed-off-by:
-
Alexander Aring authored
This patch add support to nl802154 to dump all phy capabilities which is inside the wpan_phy_supported struct. Also we introduce a new method to dumping supported channels. The new method will offer a easier interface and has lesser netlink traffic. Signed-off-by:
-
Alexander Aring authored
This patch adds the reset states for tx power levels after running reset procedure. Signed-off-by:
-
Alexander Aring authored
This patch adds reset values for cca ed level values after running reset procedure. Signed-off-by:
-
Alexander Aring authored
This patch reworks the cca energy detection level handling. This contains a calculation which works on all transceiver types and add support for dump cca energy detection levels. Signed-off-by:
-
Alexander Aring authored
This patch adds support for get transmit power levels and reworks the set of transmit power levels which was broken before. Signed-off-by:
-
Alexander Aring authored
This patch sets the at86rf230 supported cca modes. In case of at86rf212 it also can support listen before transmit. Signed-off-by:
-
Alexander Aring authored
This patch adds capability flags for supported interface types. Signed-off-by:
-
Alexander Aring authored
This patch introduce a flag property for the wpan phy structure. The current flag settings in ieee802154_hw are accessable in mac802154 layer only which is okay for flags which indicates MAC handling which are done by phy. For real PHY layer settings like cca mode, transmit power, cca energy detection level. The difference between these flags are that the MAC handling flags are only handled in mac802154/HardMac layer e.g. on an interface up. The phy settings are direct netlink calls from nl802154 into the driver layer and the nl802154 need to have a chance to check if the driver supports this handling before sending to the next layer. We also check now on PHY flags while dumping and setting pib attributes. In comparing with MIB attributes the 802.15.4 gives us an default value which we assume when a transceiver implement less functionality. In case of MIB settings the nl802154 layer doesn't need to check on the ieee802154_hw flags then. Signed-off-by:
-
Alexander Aring authored
This patch removes the check if operation is supported by driver layer. This is done now by capabilities flags, if these are valid then the driver should support the operation, otherwise a WARN_ON occurs. Signed-off-by:
-
Alexander Aring authored
This patch adds check if the value is really changed inside pib/mib. If a transceiver do support only one value for e.g. max_be then this will also handle that the driver layer doesn't need to care about handling to set one value only. Signed-off-by:
-
Alexander Aring authored
This patch adds support for phy supported handling for all other already existing handling 802.15.4 functionality. We assume now a fully 802.15.4 complaint transceiver at phy allocation. If a transceiver can support 802.15.4 default values only, then the values should be overwirtten by values the transceiver supports. If the transceiver doesn't set the according hardware flags, we assume the 802.15.4 defaults now which cannot be changed. Signed-off-by:
Phoebe Buckheister <phoebe.buckheister@itwm.fraunhofer.de> Signed-off-by:
-
Alexander Aring authored
This patch introduce the wpan_phy_supported struct for wpan_phy. There is currently no way to check if a transceiver can handle IEEE 802.15.4 complaint values. With this struct we can check before if the transceiver supports these values before sending to driver layer. Signed-off-by:
Varka Bhadram <varkabhadram@gmail.com> Cc: Alan Ott <alan@signal11.us> Signed-off-by:
-
Alexander Aring authored
This patch change the handling of cca energy detection level from dbm to mbm. This prepares to handle floating point cca energy detection levels values. The old netlink 802.15.4 will convert the dbm value to mbm for handling backward compatibility. Signed-off-by:
-
Alexander Aring authored
This patch change the handling of transmit power level from dbm to mbm. This prepares to handle floating point transmit power levels values. The old netlink 802.15.4 will convert the dbm value to mbm for handling backward compatibility. Signed-off-by:
-
Alexander Aring authored
This patch change the transmit power from s8 to s32. This prepares to store a mbm value instead dbm inside the transmit power variable. The old interface keep the a s8 dbm value, which should be backward compatibility when assign s8 to s32. Signed-off-by:
-
Alexander Aring authored
This patch moves the value validation out of softmac layer. We need to be sure now that this value is accepted by the transceiver/mac802154 or "possible" hardmac drivers before calling rdev-ops. Signed-off-by:
-
Alexander Aring authored
This patch cleanups the -EINVAL cases by combining them in one condition. Signed-off-by:
-
- 15 May, 2015 4 commits
-
-
Frederic Danis authored
Remove test of command reply status as it is already performed by __hci_cmd_sync(). __hci_cmd_sync_ev() function already returns an error if it got a non-zero status either through a Command Complete or a Command Status event. For both of these events the status is collected up in the event handlers called by hci_event_packet() and then passed as the second parameter to req_complete_skb(). The req_complete_skb() callback in turn is hci_req_sync_complete() for __hci_cmd_sync_ev() which stores the status in hdev->req_result. The hdev->req_result is then further converted through bt_to_errno() back in __hci_cmd_sync_ev(). Signed-off-by:
Frederic Danis <frederic.danis@linux.intel.com> Signed-off-by:
-
Frederic Danis authored
Remove test of command reply status as it is already performed by __hci_cmd_sync(). __hci_cmd_sync_ev() function already returns an error if it got a non-zero status either through a Command Complete or a Command Status event. For both of these events the status is collected up in the event handlers called by hci_event_packet() and then passed as the second parameter to req_complete_skb(). The req_complete_skb() callback in turn is hci_req_sync_complete() for __hci_cmd_sync_ev() which stores the status in hdev->req_result. The hdev->req_result is then further converted through bt_to_errno() back in __hci_cmd_sync_ev(). Signed-off-by:
-
Frederic Danis authored
Remove test of command reply status as it is already performed by __hci_cmd_sync(). __hci_cmd_sync_ev() function already returns an error if it got a non-zero status either through a Command Complete or a Command Status event. For both of these events the status is collected up in the event handlers called by hci_event_packet() and then passed as the second parameter to req_complete_skb(). The req_complete_skb() callback in turn is hci_req_sync_complete() for __hci_cmd_sync_ev() which stores the status in hdev->req_result. The hdev->req_result is then further converted through bt_to_errno() back in __hci_cmd_sync_ev(). Signed-off-by:
-
Frederic Danis authored
Remove test of command reply status as it is already performed by __hci_cmd_sync(). __hci_cmd_sync_ev() function already returns an error if it got a non-zero status either through a Command Complete or a Command Status event. For both of these events the status is collected up in the event handlers called by hci_event_packet() and then passed as the second parameter to req_complete_skb(). The req_complete_skb() callback in turn is hci_req_sync_complete() for __hci_cmd_sync_ev() which stores the status in hdev->req_result. The hdev->req_result is then further converted through bt_to_errno() back in __hci_cmd_sync_ev(). Signed-off-by:
-
- 14 May, 2015 1 commit
-
-
Carlo Caione authored
As already done for btintel and btbcm export setup as separate function in a vendor-specific module to hold all the Realtek specific commands. Signed-off-by:
Carlo Caione <carlo@endlessm.com> Signed-off-by:
-
- 13 May, 2015 11 commits
-
-
Xinming Hu authored
This patch fixes a compile warnning "dump_num maybe used uninitialized in this function". Signed-off-by:
Xinming Hu <huxm@marvell.com> Signed-off-by:
Amitkumar Karwar <akarwar@marvell.com> Signed-off-by:
-
Leo Yan authored
Remove the DEBUG define as the debug code; so can remove mass debug info from log buffer when using dmesg. Signed-off-by:
Leo Yan <leo.yan@linaro.org> Signed-off-by:
-
Pablo Neira authored
This fixes 4577139b ("net: use jump label patching for ingress qdisc in __netif_receive_skb_core"). The only client of this is sch_ingress and it depends on NET_CLS_ACT. So there is no way these definition can be of any help. Cc: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by:
Pablo Neira Ayuso <pablo@netfilter.org> Acked-by:
Daniel Borkmann <daniel@iogearbox.net> Acked-by:
Alexei Starovoitov <ast@plumgrid.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
David S. Miller authored
Willem de Bruijn says: ==================== refine packet socket rollover: 1. mitigate a case of lock contention 2. avoid exporting resource exhaustion to other sockets, by migrating only to a victim socket that has ample room 3. avoid reordering of most flows on the socket, by migrating first the flow responsible for load imbalance 4. help processes detect load imbalance, by exporting rollover counters Context: rollover implements flow migration in packet socket fanout groups in case of extreme load imbalance. It is a specific implementation of migration that minimizes reordering by selecting the same victim socket when possible (and by selecting subsequent victims in a round robin fashion, from which its name derives). Changes: v2 -> v3: - statistics: replace unsigned long with __aligned_u64 v1 -> v2: - huge flow detection: run lockless - huge flow detection: replace stored index with random - contention avoidance: test in packet_poll while lock held - contention avoidance: clear pressure sooner packet_poll and packet_recvmsg would clear only if the sock is empty to avoid taking the necessary lock. But, * packet_poll already holds this lock, so a lockless variant __packet_rcv_has_room is cheap. * packet_recvmsg is usually called only for non-ring sockets, which also runs lockless. - preparation: drop "single return" patch packet_rcv_has_room is now a locked wrapper around __packet_rcv_has_room, achieving the same (single footer). The benchmark mentioned in the patches is at https://github.com/wdebruij/kerneltools/blob/master/tests/bench_rollover.c ==================== Signed-off-by: -
Willem de Bruijn authored
Rollover indicates exceptional conditions. Export a counter to inform socket owners of this state. If no socket with sufficient room is found, rollover fails. Also count these events. Finally, also count when flows are rolled over early thanks to huge flow detection, to validate its correctness. Tested: Read counters in bench_rollover on all other tests in the patchset Signed-off-by:
Willem de Bruijn <willemb@google.com> Signed-off-by:
-
Willem de Bruijn authored
Migrate flows from a socket to another socket in the fanout group not only when the socket is full. Start migrating huge flows early, to divert possible 4-tuple attacks without affecting normal traffic. Introduce fanout_flow_is_huge(). This detects huge flows, which are defined as taking up more than half the load. It does so cheaply, by storing the rxhashes of the N most recent packets. If over half of these are the same rxhash as the current packet, then drop it. This only protects against 4-tuple attacks. N is chosen to fit all data in a single cache line. Tested: Ran bench_rollover for 10 sec with 1.5 Mpps of single flow input. lpbb5:/export/hda3/willemb# ./bench_rollover -l 1000 -r -s cpu rx rx.k drop.k rollover r.huge r.failed 0 14 14 0 0 0 0 1 20 20 0 0 0 0 2 16 16 0 0 0 0 3 6168824 6168824 0 4867721 4867721 0 4 4867741 4867741 0 0 0 0 5 12 12 0 0 0 0 6 15 15 0 0 0 0 7 17 17 0 0 0 0 Signed-off-by: -
Willem de Bruijn authored
Rollover has to call packet_rcv_has_room on sockets in the fanout group to find a socket to migrate to. This operation is expensive especially if the packet sockets use rings, when a lock has to be acquired. Avoid pounding on the lock by all sockets by temporarily marking a socket as "under memory pressure" when such pressure is detected. While set, only the socket owner may call packet_rcv_has_room on the socket. Once it detects normal conditions, it clears the flag. The socket is not used as a victim by any other socket in the meantime. Under reasonably balanced load, each socket writer frequently calls packet_rcv_has_room and clears its own pressure field. As a backup for when the socket is rarely written to, also clear the flag on reading (packet_recvmsg, packet_poll) if this can be done cheaply (i.e., without calling packet_rcv_has_room). This is only for edge cases. Tested: Ran bench_rollover: a process with 8 sockets in a single fanout group, each pinned to a single cpu that receives one nic recv interrupt. RPS and RFS are disabled. The benchmark uses packet rx_ring, which has to take a lock when determining whether a socket has room. Sent 3.5 Mpps of UDP traffic with sufficient entropy to spread uniformly across the packet sockets (and inserted an iptables rule to drop in PREROUTING to avoid protocol stack processing). Without this patch, all sockets try to migrate traffic to neighbors, causing lock contention when searching for a non- empty neighbor. The lock is the top 9 entries. perf record -a -g sleep 5 - 17.82% bench_rollover [kernel.kallsyms] [k] _raw_spin_lock - _raw_spin_lock - 99.00% spin_lock + 81.77% packet_rcv_has_room.isra.41 + 18.23% tpacket_rcv + 0.84% packet_rcv_has_room.isra.41 + 5.20% ksoftirqd/6 [kernel.kallsyms] [k] _raw_spin_lock + 5.15% ksoftirqd/1 [kernel.kallsyms] [k] _raw_spin_lock + 5.14% ksoftirqd/2 [kernel.kallsyms] [k] _raw_spin_lock + 5.12% ksoftirqd/7 [kernel.kallsyms] [k] _raw_spin_lock + 5.12% ksoftirqd/5 [kernel.kallsyms] [k] _raw_spin_lock + 5.10% ksoftirqd/4 [kernel.kallsyms] [k] _raw_spin_lock + 4.66% ksoftirqd/0 [kernel.kallsyms] [k] _raw_spin_lock + 4.45% ksoftirqd/3 [kernel.kallsyms] [k] _raw_spin_lock + 1.55% bench_rollover [kernel.kallsyms] [k] packet_rcv_has_room.isra.41 On net-next with this patch, this lock contention is no longer a top entry. Most time is spent in the actual read function. Next up are other locks: + 15.52% bench_rollover bench_rollover [.] reader + 4.68% swapper [kernel.kallsyms] [k] memcpy_erms + 2.77% swapper [kernel.kallsyms] [k] packet_lookup_frame.isra.51 + 2.56% ksoftirqd/1 [kernel.kallsyms] [k] memcpy_erms + 2.16% swapper [kernel.kallsyms] [k] tpacket_rcv + 1.93% swapper [kernel.kallsyms] [k] mlx4_en_process_rx_cq Looking closer at the remaining _raw_spin_lock, the cost of probing in rollover is now comparable to the cost of taking the lock later in tpacket_rcv. - 1.51% swapper [kernel.kallsyms] [k] _raw_spin_lock - _raw_spin_lock + 33.41% packet_rcv_has_room + 28.15% tpacket_rcv + 19.54% enqueue_to_backlog + 6.45% __free_pages_ok + 2.78% packet_rcv_fanout + 2.13% fanout_demux_rollover + 2.01% netif_receive_skb_internal Signed-off-by: -
Willem de Bruijn authored
Only migrate flows to sockets that have sufficient headroom, where sufficient is defined as having at least 25% empty space. The kernel has three different buffer types: a regular socket, a ring with frames (TPACKET_V[12]) or a ring with blocks (TPACKET_V3). The latter two do not expose a read pointer to the kernel, so headroom is not computed easily. All three needs a different implementation to estimate free space. Tested: Ran bench_rollover for 10 sec with 1.5 Mpps of single flow input. bench_rollover has as many sockets as there are NIC receive queues in the system. Each socket is owned by a process that is pinned to one of the receive cpus. RFS is disabled. RPS is enabled with an identity mapping (cpu x -> cpu x), to count drops with softnettop. lpbb5:/export/hda3/willemb# ./bench_rollover -r -l 1000 -s Press [Enter] to exit cpu rx rx.k drop.k rollover r.huge r.failed 0 16 16 0 0 0 0 1 21 21 0 0 0 0 2 5227502 5227502 0 0 0 0 3 18 18 0 0 0 0 4 6083289 6083289 0 5227496 0 0 5 22 22 0 0 0 0 6 21 21 0 0 0 0 7 9 9 0 0 0 0 Signed-off-by: -
Willem de Bruijn authored
Replace rollover state per fanout group with state per socket. Future patches will add fields to the new structure. Signed-off-by:
-
Willem de Bruijn authored
packet_rcv_fanout calls fanout_demux_rollover twice. Move all rollover logic into the callee to simplify these callsites, especially with upcoming changes. The main differences between the two callsites is that the FLAG variant tests whether the socket previously selected by another mode (RR, RND, HASH, ..) has room before migrating flows, whereas the rollover mode has no original socket to test. Signed-off-by:
-
Eric Dumazet authored
__ip_local_out_sk() is only used from net/ipv4/ip_output.c net/ipv4/ip_output.c:94:5: warning: symbol '__ip_local_out_sk' was not declared. Should it be static? Fixes: 7026b1dd ("netfilter: Pass socket pointer down through okfn().") Signed-off-by:
Eric Dumazet <edumazet@google.com> Signed-off-by:
-
