- 22 Jul, 2021 19 commits
-
-
Louis Peens authored
Compile the offload flow metadata and add flow_pay to the offload table. Also add in the delete paths. This does not include actual offloading to the card yet, this will follow soon. Signed-off-by:
Louis Peens <louis.peens@corigine.com> Signed-off-by:
Yinjun Zhang <yinjun.zhang@corigine.com> Signed-off-by:
Simon Horman <simon.horman@corigine.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Louis Peens authored
Combine the actions from the three different rules into one and convert into the payload format expected by the nfp. Signed-off-by:
-
Louis Peens authored
Add in the code to compile match part of the payload that will be sent to the firmware. This works similar to match.c does it, but since three flows needs to be merged it iterates through all three rules in a loop and combine the match fields to get the most strict match as result. Signed-off-by:
-
Louis Peens authored
This calculates the correct combined keylayers and key_layer_size for the to-be-offloaded flow. Signed-off-by:
-
Louis Peens authored
Change the action related offload functions to take in flow_rule * as input instead of flow_cls_offload * as input. The flow_rule parts of flow_cls_offload is the only part that is used in any case, and this is required for more conntrack offload patches which will follow later. Signed-off-by:
-
Louis Peens authored
This is a small cleanup to pass in flow->rule to some of the compile functions instead of extracting it every time. This is will also be useful for conntrack patches later. Signed-off-by:
-
Yinjun Zhang authored
Expose and refactor the match compilation functions so that they can be invoked externally. Also update the functions so they can be called multiple times with the results OR'd together. This is applicable for the flows-merging scenario, in which there could be overlapped and non-conflicting match fields. This will be used in upcoming conntrack patches. This is safe to do in the in the single call case as well since both unmasked_data and mask_data gets initialised to 0. Signed-off-by:
-
Oleksij Rempel authored
Test if we actually can send/receive packets with MTU size. This kind of issue was detected on ASIX HW with bogus EEPROM. Signed-off-by:
Oleksij Rempel <o.rempel@pengutronix.de> Signed-off-by:
-
Oleksij Rempel authored
Add missing stop and let phylib framework suspend attached PHY. Fixes: e532a096 ("net: usb: asix: ax88772: add phylib support") Signed-off-by:
-
Oleksij Rempel authored
asix_get_phyid() is used for two reasons here. To print debug message with the PHY ID and to wait until the PHY is powered up. After migrating to the phylib, we can read PHYID from sysfs. If polling for the PHY is really needed, then we will need to handle it in the phylib as well. This change was tested with: - ax88772a + internal PHY - ax88772b + external PHY Signed-off-by:
-
Vladimir Oltean authored
The newly introduced switchdev_handle_fdb_{add,del}_to_device helpers solved a problem but introduced another one. They have a severe design bug: they do not propagate FDB events on foreign interfaces to us, i.e. this use case: br0 / \ / \ / \ / \ swp0 eno0 (switchdev) (foreign) when an address is learned on eno0, what is supposed to happen is that this event should also be propagated towards swp0. Somehow I managed to convince myself that this did work correctly, but obviously it does not. The trouble with foreign interfaces is that we must reach a switchdev net_device pointer through a foreign net_device that has no direct upper/lower relationship with it. So we need to do exploratory searching through the lower interfaces of the foreign net_device's bridge upper (to reach swp0 from eno0, we must check its upper, br0, for lower interfaces that pass the check_cb and foreign_dev_check_cb). This is something that the previous code did not do, it just assumed that "dev" will become a switchdev interface at some point, somehow, probably by magic. With this patch, assisted address learning on the CPU port works again in DSA: ip link add br0 type bridge ip link set swp0 master br0 ip link set eno0 master br0 ip link set br0 up [ 46.708929] mscc_felix 0000:00:00.5 swp0: Adding FDB entry towards eno0, addr 00:04:9f:05:f4:ab vid 0 as host address Fixes: 8ca07176 ("net: switchdev: introduce a fanout helper for SWITCHDEV_FDB_{ADD,DEL}_TO_DEVICE") Reported-by:Eric Woudstra <ericwouds@gmail.com> Signed-off-by:
Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by:
-
David S. Miller authored
Vladimir Oltean says: ==================== Let switchdev drivers offload and unoffload bridge ports at their own convenience This series introduces an explicit API through which switchdev drivers mark a bridge port as offloaded or not: - switchdev_bridge_port_offload() - switchdev_bridge_port_unoffload() Currently, the bridge assumes that a port is offloaded if dev_get_port_parent_id(dev, &ppid, recurse=true) returns something, but that is just an assumption that breaks some use cases (like a non-offloaded LAG interface on top of a switchdev port, bridged with other switchdev ports). Along with some consolidation of the bridge logic to assign a "switchdev offloading mark" to a port (now better called a "hardware domain"), this series allows the bridge driver side to no longer impose restrictions on that configuration. Right now, all switchdev drivers must be modified to use the explicit API, but more and more logic can then be placed centrally in the bridge and therefore ease the job of a switchdev driver writer in the future. For example, the first thing we can hook into the explicit switchdev offloading API calls are the switchdev object and FDB replay helpers. So far, these have only been used by DSA in "pull" mode (where the driver must ask for them). Adding the replay helpers to other drivers involves a lot of repetition. But by moving the helpers inside the bridge port offload/unoffload hook points, we can move the entire replay process to "push" mode (where the bridge provides them automatically). The explicit switchdev offloading API will see further extensions in the future. The patches were split from a larger series for easier review: https://patchwork.kernel.org/project/netdevbpf/cover/20210718214434.3938850-1-vladimir.oltean@nxp.com/ Changes in v6: - Make the switchdev replay helpers opt-in - Opt out of the replay helpers for mlxsw, rocker, prestera, sparx5, cpsw, am65-cpsw ==================== Signed-off-by:
-
Vladimir Oltean authored
Starting with commit 4f2673b3 ("net: bridge: add helper to replay port and host-joined mdb entries"), DSA has introduced some bridge helpers that replay switchdev events (FDB/MDB/VLAN additions and deletions) that can be lost by the switchdev drivers in a variety of circumstances: - an IP multicast group was host-joined on the bridge itself before any switchdev port joined the bridge, leading to the host MDB entries missing in the hardware database. - during the bridge creation process, the MAC address of the bridge was added to the FDB as an entry pointing towards the bridge device itself, but with no switchdev ports being part of the bridge yet, this local FDB entry would remain unknown to the switchdev hardware database. - a VLAN/FDB/MDB was added to a bridge port that is a LAG interface, before any switchdev port joined that LAG, leading to the hardware database missing those entries. - a switchdev port left a LAG that is a bridge port, while the LAG remained part of the bridge, and all FDB/MDB/VLAN entries remained installed in the hardware database of the switchdev port. Also, since commit 0d2cfbd4 ("net: bridge: ignore switchdev events for LAG ports which didn't request replay"), DSA introduced a method, based on a const void *ctx, to ensure that two switchdev ports under the same LAG that is a bridge port do not see the same MDB/VLAN entry being replayed twice by the bridge, once for every bridge port that joins the LAG. With so many ordering corner cases being possible, it seems unreasonable to expect a switchdev driver writer to get it right from the first try. Therefore, now that DSA has experimented with the bridge replay helpers for a little bit, we can move the code to the bridge driver where it is more readily available to all switchdev drivers. To convert the switchdev object replay helpers from "pull mode" (where the driver asks for them) to a "push mode" (where the bridge offers them automatically), the biggest problem is that the bridge needs to be aware when a switchdev port joins and leaves, even when the switchdev is only indirectly a bridge port (for example when the bridge port is a LAG upper of the switchdev). Luckily, we already have a hook for that, in the form of the newly introduced switchdev_bridge_port_offload() and switchdev_bridge_port_unoffload() calls. These offer a natural place for hooking the object addition and deletion replays. Extend the above 2 functions with: - pointers to the switchdev atomic notifier (for FDB replays) and the blocking notifier (for MDB and VLAN replays). - the "const void *ctx" argument required for drivers to be able to disambiguate between which port is targeted, when multiple ports are lowers of the same LAG that is a bridge port. Most of the drivers pass NULL to this argument, except the ones that support LAG offload and have the proper context check already in place in the switchdev blocking notifier handler. Also unexport the replay helpers, since nobody except the bridge calls them directly now. Note that: (a) we abuse the terminology slightly, because FDB entries are not "switchdev objects", but we count them as objects nonetheless. With no direct way to prove it, I think they are not modeled as switchdev objects because those can only be installed by the bridge to the hardware (as opposed to FDB entries which can be propagated in the other direction too). This is merely an abuse of terms, FDB entries are replayed too, despite not being objects. (b) the bridge does not attempt to sync port attributes to newly joined ports, just the countable stuff (the objects). The reason for this is simple: no universal and symmetric way to sync and unsync them is known. For example, VLAN filtering: what to do on unsync, disable or leave it enabled? Similarly, STP state, ageing timer, etc etc. What a switchdev port does when it becomes standalone again is not really up to the bridge's competence, and the driver should deal with it. On the other hand, replaying deletions of switchdev objects can be seen a matter of cleanup and therefore be treated by the bridge, hence this patch. We make the replay helpers opt-in for drivers, because they might not bring immediate benefits for them: - nbp_vlan_init() is called _after_ netdev_master_upper_dev_link(), so br_vlan_replay() should not do anything for the new drivers on which we call it. The existing drivers where there was even a slight possibility for there to exist a VLAN on a bridge port before they join it are already guarded against this: mlxsw and prestera deny joining LAG interfaces that are members of a bridge. - br_fdb_replay() should now notify of local FDB entries, but I patched all drivers except DSA to ignore these new entries in commit 2c4eca3e ("net: bridge: switchdev: include local flag in FDB notifications"). Driver authors can lift this restriction as they wish, and when they do, they can also opt into the FDB replay functionality. - br_mdb_replay() should fix a real issue which is described in commit 4f2673b3 ("net: bridge: add helper to replay port and host-joined mdb entries"). However most drivers do not offload the SWITCHDEV_OBJ_ID_HOST_MDB to see this issue: only cpsw and am65_cpsw offload this switchdev object, and I don't completely understand the way in which they offload this switchdev object anyway. So I'll leave it up to these drivers' respective maintainers to opt into br_mdb_replay(). So most of the drivers pass NULL notifier blocks for the replay helpers, except: - dpaa2-switch which was already acked/regression-tested with the helpers enabled (and there isn't much of a downside in having them) - ocelot which already had replay logic in "pull" mode - DSA which already had replay logic in "pull" mode An important observation is that the drivers which don't currently request bridge event replays don't even have the switchdev_bridge_port_{offload,unoffload} calls placed in proper places right now. This was done to avoid unnecessary rework for drivers which might never even add support for this. For driver writers who wish to add replay support, this can be used as a tentative placement guide: https://patchwork.kernel.org/project/netdevbpf/patch/20210720134655.892334-11-vladimir.oltean@nxp.com/ Cc: Vadym Kochan <vkochan@marvell.com> Cc: Taras Chornyi <tchornyi@marvell.com> Cc: Ioana Ciornei <ioana.ciornei@nxp.com> Cc: Lars Povlsen <lars.povlsen@microchip.com> Cc: Steen Hegelund <Steen.Hegelund@microchip.com> Cc: UNGLinuxDriver@microchip.com Cc: Claudiu Manoil <claudiu.manoil@nxp.com> Cc: Alexandre Belloni <alexandre.belloni@bootlin.com> Cc: Grygorii Strashko <grygorii.strashko@ti.com> Signed-off-by:
-
Vladimir Oltean authored
There is a desire to make the object and FDB replay helpers optional when moving them inside the bridge driver. For example a certain driver might not offload host MDBs and there is no case where the replay helpers would be of immediate use to it. So it would be nice if we could allow drivers to pass NULL pointers for the atomic and blocking notifier blocks, and the replay helpers to do nothing in that case. Signed-off-by:
Florian Fainelli <f.fainelli@gmail.com> Signed-off-by:
-
Vladimir Oltean authored
On reception of an skb, the bridge checks if it was marked as 'already forwarded in hardware' (checks if skb->offload_fwd_mark == 1), and if it is, it assigns the source hardware domain of that skb based on the hardware domain of the ingress port. Then during forwarding, it enforces that the egress port must have a different hardware domain than the ingress one (this is done in nbp_switchdev_allowed_egress). Non-switchdev drivers don't report any physical switch id (neither through devlink nor .ndo_get_port_parent_id), therefore the bridge assigns them a hardware domain of 0, and packets coming from them will always have skb->offload_fwd_mark = 0. So there aren't any restrictions. Problems appear due to the fact that DSA would like to perform software fallback for bonding and team interfaces that the physical switch cannot offload. +-- br0 ---+ / / | \ / / | \ / | | bond0 / | | / \ swp0 swp1 swp2 swp3 swp4 There, it is desirable that the presence of swp3 and swp4 under a non-offloaded LAG does not preclude us from doing hardware bridging beteen swp0, swp1 and swp2. The bandwidth of the CPU is often times high enough that software bridging between {swp0,swp1,swp2} and bond0 is not impractical. But this creates an impossible paradox given the current way in which port hardware domains are assigned. When the driver receives a packet from swp0 (say, due to flooding), it must set skb->offload_fwd_mark to something. - If we set it to 0, then the bridge will forward it towards swp1, swp2 and bond0. But the switch has already forwarded it towards swp1 and swp2 (not to bond0, remember, that isn't offloaded, so as far as the switch is concerned, ports swp3 and swp4 are not looking up the FDB, and the entire bond0 is a destination that is strictly behind the CPU). But we don't want duplicated traffic towards swp1 and swp2, so it's not ok to set skb->offload_fwd_mark = 0. - If we set it to 1, then the bridge will not forward the skb towards the ports with the same switchdev mark, i.e. not to swp1, swp2 and bond0. Towards swp1 and swp2 that's ok, but towards bond0? It should have forwarded the skb there. So the real issue is that bond0 will be assigned the same hardware domain as {swp0,swp1,swp2}, because the function that assigns hardware domains to bridge ports, nbp_switchdev_add(), recurses through bond0's lower interfaces until it finds something that implements devlink (calls dev_get_port_parent_id with bool recurse = true). This is a problem because the fact that bond0 can be offloaded by swp3 and swp4 in our example is merely an assumption. A solution is to give the bridge explicit hints as to what hardware domain it should use for each port. Currently, the bridging offload is very 'silent': a driver registers a netdevice notifier, which is put on the netns's notifier chain, and which sniffs around for NETDEV_CHANGEUPPER events where the upper is a bridge, and the lower is an interface it knows about (one registered by this driver, normally). Then, from within that notifier, it does a bunch of stuff behind the bridge's back, without the bridge necessarily knowing that there's somebody offloading that port. It looks like this: ip link set swp0 master br0 | v br_add_if() calls netdev_master_upper_dev_link() | v call_netdevice_notifiers | v dsa_slave_netdevice_event | v oh, hey! it's for me! | v .port_bridge_join What we do to solve the conundrum is to be less silent, and change the switchdev drivers to present themselves to the bridge. Something like this: ip link set swp0 master br0 | v br_add_if() calls netdev_master_upper_dev_link() | v bridge: Aye! I'll use this call_netdevice_notifiers ^ ppid as the | | hardware domain for v | this port, and zero dsa_slave_netdevice_event | if I got nothing. | | v | oh, hey! it's for me! | | | v | .port_bridge_join | | | +------------------------+ switchdev_bridge_port_offload(swp0, swp0) Then stacked interfaces (like bond0 on top of swp3/swp4) would be treated differently in DSA, depending on whether we can or cannot offload them. The offload case: ip link set bond0 master br0 | v br_add_if() calls netdev_master_upper_dev_link() | v bridge: Aye! I'll use this call_netdevice_notifiers ^ ppid as the | | switchdev mark for v | bond0. dsa_slave_netdevice_event | Coincidentally (or not), | | bond0 and swp0, swp1, swp2 v | all have the same switchdev hmm, it's not quite for me, | mark now, since the ASIC but my driver has already | is able to forward towards called .port_lag_join | all these ports in hw. for it, because I have | a port with dp->lag_dev == bond0. | | | v | .port_bridge_join | for swp3 and swp4 | | | +------------------------+ switchdev_bridge_port_offload(bond0, swp3) switchdev_bridge_port_offload(bond0, swp4) And the non-offload case: ip link set bond0 master br0 | v br_add_if() calls netdev_master_upper_dev_link() | v bridge waiting: call_netdevice_notifiers ^ huh, switchdev_bridge_port_offload | | wasn't called, okay, I'll use a v | hwdom of zero for this one. dsa_slave_netdevice_event : Then packets received on swp0 will | : not be software-forwarded towards v : swp1, but they will towards bond0. it's not for me, but bond0 is an upper of swp3 and swp4, but their dp->lag_dev is NULL because they couldn't offload it. Basically we can draw the conclusion that the lowers of a bridge port can come and go, so depending on the configuration of lowers for a bridge port, it can dynamically toggle between offloaded and unoffloaded. Therefore, we need an equivalent switchdev_bridge_port_unoffload too. This patch changes the way any switchdev driver interacts with the bridge. From now on, everybody needs to call switchdev_bridge_port_offload and switchdev_bridge_port_unoffload, otherwise the bridge will treat the port as non-offloaded and allow software flooding to other ports from the same ASIC. Note that these functions lay the ground for a more complex handshake between switchdev drivers and the bridge in the future. For drivers that will request a replay of the switchdev objects when they offload and unoffload a bridge port (DSA, dpaa2-switch, ocelot), we place the call to switchdev_bridge_port_unoffload() strategically inside the NETDEV_PRECHANGEUPPER notifier's code path, and not inside NETDEV_CHANGEUPPER. This is because the switchdev object replay helpers need the netdev adjacency lists to be valid, and that is only true in NETDEV_PRECHANGEUPPER. Cc: Vadym Kochan <vkochan@marvell.com> Cc: Taras Chornyi <tchornyi@marvell.com> Cc: Ioana Ciornei <ioana.ciornei@nxp.com> Cc: Lars Povlsen <lars.povlsen@microchip.com> Cc: Steen Hegelund <Steen.Hegelund@microchip.com> Cc: UNGLinuxDriver@microchip.com Cc: Claudiu Manoil <claudiu.manoil@nxp.com> Cc: Alexandre Belloni <alexandre.belloni@bootlin.com> Cc: Grygorii Strashko <grygorii.strashko@ti.com> Signed-off-by: -
Tobias Waldekranz authored
Since hwdoms have only been used thus far for equality comparisons, the bridge has used the simplest possible assignment policy; using a counter to keep track of the last value handed out. With the upcoming transmit offloading, we need to perform set operations efficiently based on hwdoms, e.g. we want to answer questions like "has this skb been forwarded to any port within this hwdom?" Move to a bitmap-based allocation scheme that recycles hwdoms once all members leaves the bridge. This means that we can use a single unsigned long to keep track of the hwdoms that have received an skb. v1->v2: convert the typedef DECLARE_BITMAP(br_hwdom_map_t, BR_HWDOM_MAX) into a plain unsigned long. v2->v6: none Signed-off-by:Tobias Waldekranz <tobias@waldekranz.com> Signed-off-by:
-
Tobias Waldekranz authored
Before this change, four related - but distinct - concepts where named offload_fwd_mark: - skb->offload_fwd_mark: Set by the switchdev driver if the underlying hardware has already forwarded this frame to the other ports in the same hardware domain. - nbp->offload_fwd_mark: An idetifier used to group ports that share the same hardware forwarding domain. - br->offload_fwd_mark: Counter used to make sure that unique IDs are used in cases where a bridge contains ports from multiple hardware domains. - skb->cb->offload_fwd_mark: The hardware domain on which the frame ingressed and was forwarded. Introduce the term "hardware forwarding domain" ("hwdom") in the bridge to denote a set of ports with the following property: If an skb with skb->offload_fwd_mark set, is received on a port belonging to hwdom N, that frame has already been forwarded to all other ports in hwdom N. By decoupling the name from "offload_fwd_mark", we can extend the term's definition in the future - e.g. to add constraints that describe expected egress behavior - without overloading the meaning of "offload_fwd_mark". - nbp->offload_fwd_mark thus becomes nbp->hwdom. - br->offload_fwd_mark becomes br->last_hwdom. - skb->cb->offload_fwd_mark becomes skb->cb->src_hwdom. The slight change in naming here mandates a slight change in behavior of the nbp_switchdev_frame_mark() function. Previously, it only set this value in skb->cb for packets with skb->offload_fwd_mark true (ones which were forwarded in hardware). Whereas now we always track the incoming hwdom for all packets coming from a switchdev (even for the packets which weren't forwarded in hardware, such as STP BPDUs, IGMP reports etc). As all uses of skb->cb->offload_fwd_mark were already gated behind checks of skb->offload_fwd_mark, this will not introduce any functional change, but it paves the way for future changes where the ingressing hwdom must be known for frames coming from a switchdev regardless of whether they were forwarded in hardware or not (basically, if the skb comes from a switchdev, skb->cb->src_hwdom now always tracks which one). A typical example where this is relevant: the switchdev has a fixed configuration to trap STP BPDUs, but STP is not running on the bridge and the group_fwd_mask allows them to be forwarded. Say we have this setup: br0 / | \ / | \ swp0 swp1 swp2 A BPDU comes in on swp0 and is trapped to the CPU; the driver does not set skb->offload_fwd_mark. The bridge determines that the frame should be forwarded to swp{1,2}. It is imperative that forward offloading is _not_ allowed in this case, as the source hwdom is already "poisoned". Recording the source hwdom allows this case to be handled properly. v2->v3: added code comments v3->v6: none Signed-off-by:Grygorii Strashko <grygorii.strashko@ti.com> Reviewed-by:
-
Vladimir Oltean authored
Make more room for some extra code in the NETDEV_PRECHANGEUPPER handler by moving what already exists into a dedicated function. Cc: Ioana Ciornei <ioana.ciornei@nxp.com> Signed-off-by:
Ioana Ciornei <ioana.ciornei@nxp.com> Signed-off-by:
-
Vladimir Oltean authored
We need to propagate the extack argument for dpaa2_switch_port_bridge_join to use it in a future patch, and it looks like there is already an error message there which is currently printed to the console. Move it over netlink so it is properly transmitted to user space. Cc: Ioana Ciornei <ioana.ciornei@nxp.com> Signed-off-by:
-
- 21 Jul, 2021 21 commits
-
-
Leon Romanovsky authored
The failure to register devlink will leave the system with dangled devlink resource, which is not cleaned if devlink_port_register() fails. In order to remove access to ".registered" field of struct devlink_port, require both devlink_register and devlink_port_register to success and check it through device pointer. Fixes: fbfb8031 ("ionic: Add hardware init and device commands") Reviewed-by:
Moshe Shemesh <moshe@nvidia.com> Signed-off-by:
Leon Romanovsky <leonro@nvidia.com> Acked-by:
Shannon Nelson <snelson@pensando.io> Signed-off-by:
-
Nikolay Aleksandrov authored
Adding bridge multicast context support for host-joined groups is easy because we only need the proper timer value. We pass the already chosen context and use its timer value. Signed-off-by:
Nikolay Aleksandrov <nikolay@nvidia.com> Signed-off-by:
-
Nikolay Aleksandrov authored
Choose the proper bridge multicast context when user-spaces is adding mdb entries. Currently we require the vlan to be configured on at least one device (port or bridge) in order to add an mdb entry if vlan mcast snooping is enabled (vlan snooping implies vlan filtering). Note that we always allow deleting an entry, regardless of the vlan state. Signed-off-by:
-
Joakim Zhang authored
This patch fixes issues found by dtbs_check: make ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- dtbs_check DT_SCHEMA_FILES=Documentation/devicetree/bindings/net/fsl,fec.yaml According to the Micrel PHY dt-binding: Documentation/devicetree/bindings/net/micrel-ksz90x1.txt, Add clock delay in an Ethernet OF device node is deprecated, so move these properties to PHY OF device node. Suggested-by:
Rob Herring <robh@kernel.org> Signed-off-by:
Joakim Zhang <qiangqing.zhang@nxp.com> Signed-off-by:
-
Joakim Zhang authored
This patch improves the yaml a bit according to Rob Herring comments: 1) normalize interrupt-names property, there is no reason to support random order. 2) validate each string in clock-names property. 3) add constraints for fsl,num-tx-queues/fsl,num-rx-queues property. 4) change additionalProperties to false in order to do strict checking. Signed-off-by:
-
Eric Dumazet authored
tcp_grow_window() is using skb->len/skb->truesize to increase tp->rcv_ssthresh which has a direct impact on advertized window sizes. We added TCP coalescing in linux-3.4 & linux-3.5: Instead of storing skbs with one or two MSS in receive queue (or OFO queue), we try to append segments together to reduce memory overhead. High performance network drivers tend to cook skb with 3 parts : 1) sk_buff structure (256 bytes) 2) skb->head contains room to copy headers as needed, and skb_shared_info 3) page fragment(s) containing the ~1514 bytes frame (or more depending on MTU) Once coalesced into a previous skb, 1) and 2) are freed. We can therefore tweak the way we compute len/truesize ratio knowing that skb->truesize is inflated by 1) and 2) soon to be freed. This is done only for in-order skb, or skb coalesced into OFO queue. The result is that low rate flows no longer pay the memory price of having low GRO aggregation factor. Same result for drivers not using GRO. This is critical to allow a big enough receiver window, typically tcp_rmem[2] / 2. We have been using this at Google for about 5 years, it is due time to make it upstream. Signed-off-by:
Eric Dumazet <edumazet@google.com> Cc: Soheil Hassas Yeganeh <soheil@google.com> Cc: Neal Cardwell <ncardwell@google.com> Cc: Yuchung Cheng <ycheng@google.com> Acked-by:
Soheil Hassas Yeganeh <soheil@google.com> Signed-off-by:
-
Nikolay Aleksandrov authored
With the recent change to use bridge/port multicast context pointers instead of bridge/port I missed to convert two locations which pass the port pointer as-is, but with the new model we need to verify the port context is non-NULL first and retrieve the port from it. The first location is when doing querier selection when a query is received, the second location is when leaving a group. The port context will be null if the packets originated from the bridge device (i.e. from the host). The fix is simple just check if the port context exists and retrieve the port pointer from it. Fixes: adc47037 ("net: bridge: multicast: use multicast contexts instead of bridge or port") Signed-off-by:
-
Leon Romanovsky authored
The driver core will call to .remove callback only if .probe succeeded and it will ensure that driver data has pointer to struct ionic. There is no need to check it again. Fixes: fbfb8031 ("ionic: Add hardware init and device commands") Signed-off-by:
-
Eric Dumazet authored
For tcp sockets, sk->sk_write_space is most probably sk_stream_write_space(). Other sk->sk_write_space() calls in TCP are slow path and do not deserve any change. Signed-off-by:
-
Andy Shevchenko authored
Eliminate some boilerplate code by using module_pci_driver() instead of init/exit, moving the salient bits from init into probe. Signed-off-by:
Andy Shevchenko <andriy.shevchenko@linux.intel.com> Reviewed-by:
Loic Poulain <loic.poulain@linaro.org> Signed-off-by:
-
Dongliang Mu authored
There are two invocation sites of hso_free_net_device. After refactoring hso_create_net_device, this parameter is useless. Remove the bailout in the hso_free_net_device and change the invocation sites of this function. Signed-off-by:
Dongliang Mu <mudongliangabcd@gmail.com> Signed-off-by:
-
Dongliang Mu authored
The current error handling code of hso_create_net_device is hso_free_net_device, no matter which errors lead to. For example, WARNING in hso_free_net_device [1]. Fix this by refactoring the error handling code of hso_create_net_device by handling different errors by different code. [1] https://syzkaller.appspot.com/bug?id=66eff8d49af1b28370ad342787413e35bbe76efe Reported-by: syzbot+44d53c7255bb1aea22d2@syzkaller.appspotmail.com Fixes: 5fcfb6d0 ("hso: fix bailout in error case of probe") Signed-off-by:
-
Piotr Kwapulinski authored
Add support for external synchronization clock via GPIOs. 1PPS signals are handled via the dedicated 3 GPIOs: SDP3_2, SDP3_3 and GPIO_4. Previously it was not possible to use the external PTP synchronization clock. All possible HW configurations are supported. SDP3_2, SDP3_3, GPIO_4 off, off, off off, in_A, off off, out_A, off off, in_B, off off, out_B, off in_A, off, off in_A, in_B, off in_A, out_B, off out_A, off, off out_A, in_B, off in_B, off, off in_B, in_A, off in_B, out_A, off out_B, off, off out_B, in_A, off off, off, in_A off, out_A, in_A off, in_B, in_A off, out_B, in_A out_A, off, in_A out_A, in_B, in_A in_B, off, in_A in_B, out_A, in_A out_B, off, in_A off, off, out_A off, in_A, out_A off, in_B, out_A off, out_B, out_A in_A, off, out_A in_A, in_B, out_A in_A, out_B, out_A in_B, off, out_A in_B, in_A, out_A out_B, off, out_A out_B, in_A, out_A off, off, in_B off, in_A, in_B off, out_A, in_B off, out_B, in_B in_A, off, in_B in_A, out_B, in_B out_A, off, in_B out_B, off, in_B out_B, in_A, in_B off, off, out_B off, in_A, out_B off, out_A, out_B off, in_B, out_B in_A, off, out_B in_A, in_B, out_B out_A, off, out_B out_A, in_B, out_B in_B, off, out_B in_B, in_A, out_B in_B, out_A, out_B Tested with oscilloscope, 1PPS generator and ts2phc. Reviewed-by:
Aleksandr Loktionov <aleksandr.loktionov@intel.com> Reviewed-by:
Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com> Signed-off-by:
Piotr Kwapulinski <piotr.kwapulinski@intel.com> Tested-by:
Ashish K <ashishx.k@intel.com> Signed-off-by:
Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by:
-
Vadim Fedorenko authored
Consolidate IPv4 MTU code the same way it is done in IPv6 to have code aligned in both address families Signed-off-by:
Vadim Fedorenko <vfedorenko@novek.ru> Reviewed-by:
David Ahern <dsahern@kernel.org> Signed-off-by:
-
Vadim Fedorenko authored
Replace ip6_dst_mtu_forward with ip6_dst_mtu_maybe_forward and reuse this code in ip6_mtu. Actually these two functions were almost duplicates, this change will simplify the maintaince of mtu calculation code. Signed-off-by:
-
David S. Miller authored
Justin Iurman says: ==================== Support for the IOAM Pre-allocated Trace with IPv6 v5: - Refine types, min/max and default values for new sysctls - Introduce a "_wide" sysctl for each "ioam6_id" sysctl - Add more validation on headers before processing data - RCU for sc <> ns pointers + appropriate accessors - Generic Netlink policies are now per op, not per family anymore - Address other comments/remarks from Jakub (thanks again) - Revert "__packed" to "__attribute__((packed))" for uapi headers - Add tests to cover the functionality added, as requested by David Ahern v4: - Address warnings from checkpatch (ignore errors related to unnamed bitfields in the first patch) - Use of hweight32 (thanks Jakub) - Remove inline keyword from static functions in C files and let the compiler decide what to do (thanks Jakub) v3: - Fix warning "unused label 'out_unregister_genl'" by adding conditional macro - Fix lwtunnel output redirect bug: dst cache useless in this case, use orig_output instead v2: - Fix warning with static for __ioam6_fill_trace_data - Fix sparse warning with __force when casting __be64 to __be32 - Fix unchecked dereference when removing IOAM namespaces or schemas - exthdrs.c: Don't drop by default (now: ignore) to match the act bits "00" - Add control plane support for the inline insertion (lwtunnel) - Provide uapi structures - Use __net_timestamp if skb->tstamp is empty - Add note about the temporary IANA allocation - Remove support for "removable" TLVs - Remove support for virtual/anonymous tunnel decapsulation In-situ Operations, Administration, and Maintenance (IOAM) records operational and telemetry information in a packet while it traverses a path between two points in an IOAM domain. It is defined in draft-ietf-ippm-ioam-data [1]. IOAM data fields can be encapsulated into a variety of protocols. The IPv6 encapsulation is defined in draft-ietf-ippm-ioam-ipv6-options [2], via extension headers. IOAM can be used to complement OAM mechanisms based on e.g. ICMP or other types of probe packets. This patchset implements support for the Pre-allocated Trace, carried by a Hop-by-Hop. Therefore, a new IPv6 Hop-by-Hop TLV option is introduced, see IANA [3]. The three other IOAM options are not included in this patchset (Incremental Trace, Proof-of-Transit and Edge-to-Edge). The main idea behind the IOAM Pre-allocated Trace is that a node pre-allocates some room in packets for IOAM data. Then, each IOAM node on the path will insert its data. There exist several interesting use- cases, e.g. Fast failure detection/isolation or Smart service selection. Another killer use-case is what we have called Cross-Layer Telemetry, see the demo video on its repository [4], that aims to make the entire stack (L2/L3 -> L7) visible for distributed tracing tools (e.g. Jaeger), instead of the current L5 -> L7 limited view. So, basically, this is a nice feature for the Linux Kernel. This patchset also provides support for the control plane part, but only for the inline insertion (host-to-host use case), through lightweight tunnels. Indeed, for in-transit traffic, the solution is to have an IPv6-in-IPv6 encapsulation, which brings some difficulties and still requires a little bit of work and discussion (ie anonymous tunnel decapsulation and multi egress resolution). - Patch 1: IPv6 IOAM headers definition - Patch 2: Data plane support for Pre-allocated Trace - Patch 3: IOAM Generic Netlink API - Patch 4: Support for IOAM injection with lwtunnels - Patch 5: Documentation for new IOAM sysctls - Patch 6: Test for the IOAM insertion with IPv6 [1] https://tools.ietf.org/html/draft-ietf-ippm-ioam-data [2] https://tools.ietf.org/html/draft-ietf-ippm-ioam-ipv6-options [3] https://www.iana.org/assignments/ipv6-parameters/ipv6-parameters.xhtml#ipv6-parameters-2 [4] https://github.com/iurmanj/cross-layer-telemetry ==================== Signed-off-by:
-
Justin Iurman authored
This test evaluates the IOAM insertion for IPv6 by checking the IOAM data integrity on the receiver. The topology is formed by 3 nodes: Alpha (sender), Beta (router in-between) and Gamma (receiver). An IOAM domain is configured from Alpha to Gamma only, which means not on the reverse path. When Gamma is the destination, Alpha adds an IOAM option (Pre-allocated Trace) inside a Hop-by-hop and fills the trace with its own IOAM data. Beta and Gamma also fill the trace. The IOAM data integrity is checked on Gamma, by comparing with the pre-defined IOAM configuration (see below). +-------------------+ +-------------------+ | | | | | alpha netns | | gamma netns | | | | | | +-------------+ | | +-------------+ | | | veth0 | | | | veth0 | | | | db01::2/64 | | | | db02::2/64 | | | +-------------+ | | +-------------+ | | . | | . | +-------------------+ +-------------------+ . . . . . . +----------------------------------------------------+ | . . | | +-------------+ +-------------+ | | | veth0 | | veth1 | | | | db01::1/64 | ................ | db02::1/64 | | | +-------------+ +-------------+ | | | | beta netns | | | +--------------------------+-------------------------+ ~~~~~~~~~~~~~~~~~~~~~~ | IOAM configuration | ~~~~~~~~~~~~~~~~~~~~~~ Alpha +-----------------------------------------------------------+ | Type | Value | +-----------------------------------------------------------+ | Node ID | 1 | +-----------------------------------------------------------+ | Node Wide ID | 11111111 | +-----------------------------------------------------------+ | Ingress ID | 0xffff (default value) | +-----------------------------------------------------------+ | Ingress Wide ID | 0xffffffff (default value) | +-----------------------------------------------------------+ | Egress ID | 101 | +-----------------------------------------------------------+ | Egress Wide ID | 101101 | +-----------------------------------------------------------+ | Namespace Data | 0xdeadbee0 | +-----------------------------------------------------------+ | Namespace Wide Data | 0xcafec0caf00dc0de | +-----------------------------------------------------------+ | Schema ID | 777 | +-----------------------------------------------------------+ | Schema Data | something that will be 4n-aligned | +-----------------------------------------------------------+ Note: When Gamma is the destination, Alpha adds an IOAM Pre-allocated Trace option inside a Hop-by-hop, where 164 bytes are pre-allocated for the trace, with 123 as the IOAM-Namespace and with 0xfff00200 as the trace type (= all available options at this time). As a result, and based on IOAM configurations here, only both Alpha and Beta should be capable of inserting their IOAM data while Gamma won't have enough space and will set the overflow bit. Beta +-----------------------------------------------------------+ | Type | Value | +-----------------------------------------------------------+ | Node ID | 2 | +-----------------------------------------------------------+ | Node Wide ID | 22222222 | +-----------------------------------------------------------+ | Ingress ID | 201 | +-----------------------------------------------------------+ | Ingress Wide ID | 201201 | +-----------------------------------------------------------+ | Egress ID | 202 | +-----------------------------------------------------------+ | Egress Wide ID | 202202 | +-----------------------------------------------------------+ | Namespace Data | 0xdeadbee1 | +-----------------------------------------------------------+ | Namespace Wide Data | 0xcafec0caf11dc0de | +-----------------------------------------------------------+ | Schema ID | 0xffffff (= None) | +-----------------------------------------------------------+ | Schema Data | | +-----------------------------------------------------------+ Gamma +-----------------------------------------------------------+ | Type | Value | +-----------------------------------------------------------+ | Node ID | 3 | +-----------------------------------------------------------+ | Node Wide ID | 33333333 | +-----------------------------------------------------------+ | Ingress ID | 301 | +-----------------------------------------------------------+ | Ingress Wide ID | 301301 | +-----------------------------------------------------------+ | Egress ID | 0xffff (default value) | +-----------------------------------------------------------+ | Egress Wide ID | 0xffffffff (default value) | +-----------------------------------------------------------+ | Namespace Data | 0xdeadbee2 | +-----------------------------------------------------------+ | Namespace Wide Data | 0xcafec0caf22dc0de | +-----------------------------------------------------------+ | Schema ID | 0xffffff (= None) | +-----------------------------------------------------------+ | Schema Data | | +-----------------------------------------------------------+ Signed-off-by:Justin Iurman <justin.iurman@uliege.be> Signed-off-by:
-
Justin Iurman authored
Add documentation for new IOAM sysctls: - ioam6_id and ioam6_id_wide: two per-namespace sysctls - ioam6_enabled, ioam6_id and ioam6_id_wide: three per-interface sysctls Example of IOAM configuration based on the following simple topology: _____ _____ _____ | | eth0 eth0 | | eth1 eth0 | | | A |.----------.| B |.----------.| C | |_____| |_____| |_____| 1) Node and interface IDs can be configured for IOAM: # IOAM ID of A = 1, IOAM ID of A.eth0 = 11 (A) sysctl -w net.ipv6.ioam6_id=1 (A) sysctl -w net.ipv6.conf.eth0.ioam6_id=11 # IOAM ID of B = 2, IOAM ID of B.eth0 = 21, IOAM ID of B.eth1 = 22 (B) sysctl -w net.ipv6.ioam6_id=2 (B) sysctl -w net.ipv6.conf.eth0.ioam6_id=21 (B) sysctl -w net.ipv6.conf.eth1.ioam6_id=22 # IOAM ID of C = 3, IOAM ID of C.eth0 = 31 (C) sysctl -w net.ipv6.ioam6_id=3 (C) sysctl -w net.ipv6.conf.eth0.ioam6_id=31 Note that "_wide" IDs equivalents can be configured the same way. 2) Each node can be configured to form an IOAM domain. For instance, we allow IOAM from A to C only (not the reverse path), i.e. enable IOAM on ingress for B.eth0 and C.eth0: (B) sysctl -w net.ipv6.conf.eth0.ioam6_enabled=1 (C) sysctl -w net.ipv6.conf.eth0.ioam6_enabled=1 3) An IOAM domain (e.g. ID=123) is defined and made known to each node: (A) ip ioam namespace add 123 (B) ip ioam namespace add 123 (C) ip ioam namespace add 123 4) Finally, an IOAM Pre-allocated Trace can be inserted in traffic sent by A when C (e.g. db02::2) is the destination: (A) ip -6 route add db02::2/128 encap ioam6 trace type 0x800000 ns 123 size 12 dev eth0 Signed-off-by: -
Justin Iurman authored
Add support for the IOAM inline insertion (only for the host-to-host use case) which is per-route configured with lightweight tunnels. The target is iproute2 and the patch is ready. It will be posted as soon as this patchset is merged. Here is an overview: $ ip -6 ro ad fc00::1/128 encap ioam6 trace type 0x800000 ns 1 size 12 dev eth0 This example configures an IOAM Pre-allocated Trace option attached to the fc00::1/128 prefix. The IOAM namespace (ns) is 1, the size of the pre-allocated trace data block is 12 octets (size) and only the first IOAM data (bit 0: hop_limit + node id) is included in the trace (type) represented as a bitfield. The reason why the in-transit (IPv6-in-IPv6 encapsulation) use case is not implemented is explained on the patchset cover. Signed-off-by:
-
Justin Iurman authored
Add Generic Netlink commands to allow userspace to configure IOAM namespaces and schemas. The target is iproute2 and the patch is ready. It will be posted as soon as this patchset is merged. Here is an overview: $ ip ioam Usage: ip ioam { COMMAND | help } ip ioam namespace show ip ioam namespace add ID [ data DATA32 ] [ wide DATA64 ] ip ioam namespace del ID ip ioam schema show ip ioam schema add ID DATA ip ioam schema del ID ip ioam namespace set ID schema { ID | none } Signed-off-by: -
Justin Iurman authored
Implement support for processing the IOAM Pre-allocated Trace with IPv6, see [1] and [2]. Introduce a new IPv6 Hop-by-Hop TLV option, see IANA [3]. A new per-interface sysctl is introduced. The value is a boolean to accept (=1) or ignore (=0, by default) IPv6 IOAM options on ingress for an interface: - net.ipv6.conf.XXX.ioam6_enabled Two other sysctls are introduced to define IOAM IDs, represented by an integer. They are respectively per-namespace and per-interface: - net.ipv6.ioam6_id - net.ipv6.conf.XXX.ioam6_id The value of the first one represents the IOAM ID of the node itself (u32; max and default value = U32_MAX>>8, due to hop limit concatenation) while the other represents the IOAM ID of an interface (u16; max and default value = U16_MAX). Each "ioam6_id" sysctl has a "_wide" equivalent: - net.ipv6.ioam6_id_wide - net.ipv6.conf.XXX.ioam6_id_wide The value of the first one represents the wide IOAM ID of the node itself (u64; max and default value = U64_MAX>>8, due to hop limit concatenation) while the other represents the wide IOAM ID of an interface (u32; max and default value = U32_MAX). The use of short and wide equivalents is not exclusive, a deployment could choose to leverage both. For example, net.ipv6.conf.XXX.ioam6_id (short format) could be an identifier for a physical interface, whereas net.ipv6.conf.XXX.ioam6_id_wide (wide format) could be an identifier for a logical sub-interface. Documentation about new sysctls is provided at the end of this patchset. Two relativistic hash tables are used: one for IOAM namespaces, the other for IOAM schemas. A namespace can only have a single active schema and a schema can only be attached to a single namespace (1:1 relationship). [1] https://tools.ietf.org/html/draft-ietf-ippm-ioam-ipv6-options [2] https://tools.ietf.org/html/draft-ietf-ippm-ioam-data [3] https://www.iana.org/assignments/ipv6-parameters/ipv6-parameters.xhtml#ipv6-parameters-2Signed-off-by:
-
