An error occurred fetching the project authors.
- 14 Jul, 2020 5 commits
-
-
Petr Machata authored
The RED qevents early_drop and mark can be offloaded under the following fairly strict conditions: - At most one filter is configured at the qevent block - The protocol is "any" - The classifier is matchall - The action is trap, sample, or mirror with the same conditions as with other SPAN offloads - The hw_counters type is none In this patchset, implement offload of mirror for early_drop qevent. The ECN trigger is currently not implemented in the FW and therefore the mark qevent is not supported. The qevent notifications look exactly like regular block binding notifications with a binder type that identifies them as qevents. Therefore the details of processing this binding are fairly similar to the matchall offload. struct flow_block_offload.sch points at the qdisc in question. Use it to figure out if the qdisc is offloaded at all and what TC it configures. Bounce bindings on not-offloaded qdiscs. Individual bindings are kept in a list so that several qevents can share the same block and all binding points get configured as the configured filters change. Signed-off-by:
Petr Machata <petrm@mellanox.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Petr Machata authored
Two RED qevents have been introduced recently. From the point of view of a driver, qevents are simply blocks with unusual binder types. However they need to be handled by different logic than ACL-like flows. Thus rename mlxsw_sp_setup_tc_block() to mlxsw_sp_setup_tc_block_clsact() and move the binder-type dispatch from there to spectrum.c into a new function of the original name. The new dispatcher is easier to extend with new binder types. Signed-off-by:
Jiri Pirko <jiri@mellanox.com> Signed-off-by:
-
Petr Machata authored
A following patch introduces offloading of filters attached to blocks bound to the RED tail_drop qevent. The only classifier that mlxsw will permit in this role is matchall. mlxsw currently offloads matchall filters used with clsact qdisc. The data structures used for that offload will come handy for the qevent offload as well. Publish them in spectrum.h. Signed-off-by:
-
Petr Machata authored
The field "dev" in struct mlxsw_sp_flow_block_binding is not used. Drop it. Signed-off-by:
-
Ido Schimmel authored
The per-ASIC SPAN operations are relevant to the SPAN module and therefore should be implemented there and not in the main driver file. Move them. These operations will be extended later on. Signed-off-by:
Ido Schimmel <idosch@mellanox.com> Reviewed-by:
-
- 30 Jun, 2020 2 commits
-
-
Amit Cohen authored
Move mlxsw_sp1_port_type_speed_ops and mlxsw_sp2_port_type_speed_ops with the relevant code from spectrum.c to spectrum_ethtool.c. Signed-off-by:
Amit Cohen <amitc@mellanox.com> Reviewed-by:
Florian Fainelli <f.fainelli@gmail.com> Signed-off-by:
-
Amit Cohen authored
Add spectrum_ethtool.c file for ethtool code. Move ethtool_ops and the relevant code from spectrum.c to spectrum_ethtool.c. Signed-off-by:
-
- 22 Jun, 2020 2 commits
-
-
Petr Machata authored
Certain ACL actions are only available on some Spectrum revisions. In particular, L4_PORT_ACTION is not available on Spectrum-1. Introduce a new ops struct intended to hold these differences, mlxsw_sp_rulei_ops. Prime it with a sole member, act_mangle_field, meant for handling of pedit mangles. Create two ops structures, one for Spectrum-1, the other for Spectrum-2 and above. Add callbacks for act_mangle_field and dispatch to the common handler. Invoke mlxsw_sp_rulei_ops.act_mangle_field from the field mangler instead of calling the common handler directly. Signed-off-by:
-
Ido Schimmel authored
The second commit cited below performed a cast of 'u32 buffsize' to '(u16 *)' when calling mlxsw_sp_port_headroom_8x_adjust(): mlxsw_sp_port_headroom_8x_adjust(mlxsw_sp_port, (u16 *) &buffsize); Colin noted that this will behave differently on big endian architectures compared to little endian architectures. Fix this by following Colin's suggestion and have the function accept and return 'u32' instead of passing the current size by reference. Fixes: da382875 ("mlxsw: spectrum: Extend to support Spectrum-3 ASIC") Fixes: 60833d54 ("mlxsw: spectrum: Adjust headroom buffers for 8x ports") Signed-off-by:
Colin Ian King <colin.king@canonical.com> Suggested-by:
-
- 16 Jun, 2020 1 commit
-
-
Ido Schimmel authored
The port's headroom buffers are used to store packets while they traverse the device's pipeline and also to store packets that are egress mirrored. On Spectrum-3, ports with eight lanes use two headroom buffers between which the configured headroom size is split. In order to prevent packet loss, multiply the calculated headroom size by two for 8x ports. Fixes: da382875 ("mlxsw: spectrum: Extend to support Spectrum-3 ASIC") Signed-off-by:
-
- 01 Jun, 2020 2 commits
-
-
Ido Schimmel authored
In a similar fashion to other control traps, register ACL control traps with devlink. Signed-off-by:
-
Ido Schimmel authored
In a similar fashion to other traps, register layer 2 control traps with devlink. Signed-off-by:
-
- 09 May, 2020 4 commits
-
-
Jiri Pirko authored
On ingress, the matchall rules doing mirroring and sampling are offloaded into hardware blocks that are processed before any flower rules. On egress, the matchall mirroring rules are offloaded into hardware block that is processed after all flower rules. Therefore check the priorities of inserted matchall rules against existing flower rules and ensure the correct ordering. Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Jiri Pirko authored
Introduce an infrastructure that allows to get minimum and maximum rule priority for specified chain. This is going to be used by a subsequent patch to enforce ordering between flower and matchall filters. Signed-off-by:
-
Jiri Pirko authored
As there are going to be other matchall specific fields in flow structure, put the existing list field into matchall substruct. Signed-off-by:
-
Jiri Pirko authored
Introduce an infrastructure that allows to get minimum and maximum rule priority for specified chain. This is going to be used by a subsequent patch to enforce ordering between flower and matchall filters. Signed-off-by:
-
- 27 Apr, 2020 7 commits
-
-
Jiri Pirko authored
Move the code taking case of setup of flow offload into spectrum_flow.c Do small renaming of callbacks on the way. Signed-off-by:
-
Jiri Pirko authored
Currently there are two callbacks registered: one for matchall, one for flower. This causes the user to see "in_hw_count 2" in TC filter dump. Because of this and also as a preparation for future matchall offload for rules equivalent to flower-all-match, move the processing of shared block into matchall.c. Leave only one cb for mlxsw driver per-block. Signed-off-by:
-
Jiri Pirko authored
Currently, only the psample_group is accessed using RCU on RX path. However, it is possible (unlikely) that other sample values get change during RX processing. Fix this by having the port->sample struct accessed as RCU pointer, containing all sample values including psample_group pointer. That avoids extra alloc per-port, copying the values and the race condition described above. Signed-off-by:
-
Jiri Pirko authored
Similar to flower, have matchall related code in a separate file. Do some small renaming on the way (consistent "mall" prefixes, dropped "_tc_", dropped "_port_" where suitable). Signed-off-by:
-
Jiri Pirko authored
The code around flow_block is currently mixed in spectrum_acl.c. However, as it really does not directly relate to ACL part only, push the bits into a separate file. Signed-off-by:
-
Jiri Pirko authored
The acl_block structure is going to be used for non-acl case - matchall offload. So rename it accordingly. Signed-off-by:
-
Jiri Pirko authored
The struct is defined in the header, no need to have the helpers in the c file. Move the helpers to the header. Signed-off-by:
-
- 31 Mar, 2020 3 commits
-
-
Ido Schimmel authored
Implement support for setting of packet trap group parameters by invoking the trap_group_init() callback with the new parameters. Signed-off-by:
-
Ido Schimmel authored
Register supported packet trap policers with devlink and implement callbacks to change their parameters and read their counters. Prevent user space from passing invalid policer parameters down to the device by checking their validity and communicating the failure via an appropriate extack message. v2: * Remove the max/min validity checks from __mlxsw_sp_trap_policer_set() Signed-off-by:
-
Ido Schimmel authored
During initialization the driver configures various packet trap groups and binds policers to them. Currently, most of these groups are not exposed to user space and therefore their policers should not be exposed as well. Otherwise, user space will be able to alter policer parameters without knowing which packet traps are policed by the policer. Use a bitmap to track the used policer IDs so that these policers will not be registered with devlink in a subsequent patch. Signed-off-by:
-
- 30 Mar, 2020 1 commit
-
-
Jiri Pirko authored
It may be up to the driver (in case ANY HW stats is passed) to select which type of HW stats he is going to use. Add an infrastructure to expose this information to user. $ tc filter add dev enp3s0np1 ingress proto ip handle 1 pref 1 flower dst_ip 192.168.1.1 action drop $ tc -s filter show dev enp3s0np1 ingress filter protocol ip pref 1 flower chain 0 filter protocol ip pref 1 flower chain 0 handle 0x1 eth_type ipv4 dst_ip 192.168.1.1 in_hw in_hw_count 2 action order 1: gact action drop random type none pass val 0 index 1 ref 1 bind 1 installed 10 sec used 10 sec Action statistics: Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0 used_hw_stats immediate <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Signed-off-by:
-
- 26 Mar, 2020 1 commit
-
-
Petr Machata authored
Offload action pedit ex munge when used with a flower classifier. Only allow setting of DSCP, ECN, or the whole DSField in IPv4 and IPv6 packets. Signed-off-by:
-
- 20 Mar, 2020 1 commit
-
-
Petr Machata authored
Offload action skbedit priority when keyed to a flower classifier. The skb->priority field in Linux is very generic, so only allow setting the bottom 8 priorities and bounce anything else. Signed-off-by:
-
- 18 Mar, 2020 1 commit
-
-
Jiri Pirko authored
Implement devlink resources support for counter pools. Move the subpool sizes calculations into the new resources register function. Signed-off-by:
-
- 09 Mar, 2020 1 commit
-
-
Jiri Pirko authored
Set a flag in case rule counter was created. Only query the device for stats of a rule, which has the valid counter assigned. Signed-off-by:
-
- 05 Mar, 2020 2 commits
-
-
Petr Machata authored
There are two peculiarities about offloading FIFO: - sometimes the qdisc has an unspecified handle (it is "invisible") - it may be created before the qdisc that it will be a child of These features make the offload a bit more tricky. The approach chosen in this patch is to make note of all the FIFOs that needed to be rejected because their parents were not known. Later when the parent is created, they are offloaded FIFO is only offloaded for its counters, queue length is ignored. Signed-off-by:
-
Petr Machata authored
In order to have a tidy structure where to put information related to Qdisc offloads, introduce a new structure. Move there the two existing pieces of data: root_qdisc and tclass_qdiscs. Embed them directly, because there's no reason to go through pointer anymore. Convert users, update init/fini functions. Signed-off-by:
-
- 27 Feb, 2020 1 commit
-
-
Ido Schimmel authored
During port initialization the driver instructs the device to only advertise speeds that can be supported by the port's current width. Since the device now returns the supported speeds based on the port's current width, the driver no longer needs to compute the speeds that can be advertised. Simplify port initialization by setting the advertised speeds to the queried supported speeds. Signed-off-by:
-
- 25 Feb, 2020 2 commits
-
-
Jiri Pirko authored
Use the cookie index received along with the packet to lookup original flow_offload cookie binary and pass it down to devlink_trap_report(). Add "fa_cookie" metadata to the ACL trap. Signed-off-by:
-
Jiri Pirko authored
Track cookies coming down to driver by flow_offload. Assign a cookie_index to each unique cookie binary. Use previously defined "Trap with userdef" flex action to ask HW to pass cookie_index alongside with the dropped packets. Signed-off-by:
-
- 24 Feb, 2020 4 commits
-
-
Jiri Pirko authored
The ACL flex action will have to know if it is in ingress or egress, so it can use correct trap ID. Pass the ingress indication down to it. Signed-off-by:
-
Jiri Pirko authored
Action drop is going to be tracked by two separate traps, one for ingress and one for egress. Prepare for it and disallow the possibility to have drop action in blocks which are bound to both ingress and egress. Signed-off-by:
-
Jiri Pirko authored
Count the number of ingress and egress block bindings. Use the egress counter in "is_egress_bound" helper. Add couple of helpers to check ingress and mixed bound. Signed-off-by:
-
Jiri Pirko authored
There are couple of places where block pointer as a function argument can be const. So make those const. Signed-off-by:
-
