summaryrefslogtreecommitdiff
path: root/drivers
AgeCommit message (Collapse)Author
24 hoursnet/mlx5: set maximum LRO packet sizeDekel Peled
This patch implements use of the API for LRO aggregated packet max size. Rx queue create is updated to use the relevant configuration. Documentation is updated accordingly. Signed-off-by: Dekel Peled <dekelp@mellanox.com> Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
32 hoursnet/dpaa2: fix Rx offload flags on jumbo MTU setSachin Saxena
The JUMBO frame handling in dpaa2_dev_mtu_set api was not correct. When frame_size is greater than RTE_ETHER_MAX_LEN, the intention is to add JUMBO flag in rx offload while it was resetting all other flags other than JUMBO as AND operator was used instead of OR. Fixes: 0ebce6129bc6 ("net/dpaa2: support new ethdev offload APIs") Cc: stable@dpdk.org Signed-off-by: Sachin Saxena <sachin.saxena@nxp.com> Reviewed-by: Gagandeep Singh <g.singh@nxp.com>
33 hoursnet/mlx5: set VF MAC address from hostRaslan Darawsheh
Allow to configure the default MAC address of a VF via its representor port in the host. An API was proposed to specify explicitly the VF as a target: https://patches.dpdk.org/patch/62176/ It has been rejected by the technical board in order to keep compatibility with behavior in Intel PMDs. http://mails.dpdk.org/archives/dev/2019-November/150588.html Signed-off-by: Thomas Monjalon <thomas@monjalon.net> Signed-off-by: Raslan Darawsheh <rasland@mellanox.com> Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
33 hoursdrivers/net: update Rx RSS hash offload capabilitiesPavan Nikhilesh
Add DEV_RX_OFFLOAD_RSS_HASH flag for all PMDs that support RSS hash delivery. Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com> Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com> Reviewed-by: Hemant Agrawal <hemant.agrawal@nxp.com> Acked-by: Jerin Jacob <jerinj@marvell.com> Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
35 hoursdoc: add ice PMD in doxygenHaiyue Wang
Add the doxygen for ice protocol extraction feature APIs. Signed-off-by: Haiyue Wang <haiyue.wang@intel.com> Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
35 hoursnet/ice: optimize protocol extraction by dynamic mbufHaiyue Wang
The original design is to use rte_mbuf::udata64 to save the metadata of protocol extraction which has network protocol data fields and type, a private API is used to decode this metadata. Use the dynamic mbuf field and flags to register the needed fields in mbuf, to avoid overwriting 'rte_mbuf::udata64', since the application may use it. Now the protocol extraction metadate is saved into dynamic mbuf field with 4B size, and its type and validity is indicated by the related dynamic mbuf flags in 'rte_mbuf::ol_flags'. Signed-off-by: Haiyue Wang <haiyue.wang@intel.com> Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
35 hoursnet/mlx5: optimize tag traversal with hash listBing Zhao
Tag action for flow mark/flag could be reused by different flows. When creating a new flow with mark, the existing tag resources will be traversed in order to confirm if the action is already created. If only one linked list is used, the searching rate will drop significantly with the number of tag actions increasing. By using a hash lists table, it will speed up the searching process and in the meanwhile, the memory consumption won't be large if only a small number tag action resources are created(compared to other hash table implementations). The list heads array size could be optimized with some extendable hash table in the future. Signed-off-by: Bing Zhao <bingz@mellanox.com> Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
35 hoursnet/mlx5: reorganize flow matcher resourcesBing Zhao
Matchers are created on the specific table. If a single linked list is used to store these, then the finding process might be the bottleneck when there are a lot of different flow matchers on a huge amount of tables. The matchers could be move into the table data resource structure in order to reduce the comparison times when finding. Signed-off-by: Bing Zhao <bingz@mellanox.com> Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
35 hoursnet/mlx5: reorganize jump table resourcesBing Zhao
Jump object is associated with table object, so there is no need to use a single linked list to store it. All the jump objects could be put together with related flow tables. Signed-off-by: Bing Zhao <bingz@mellanox.com> Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
35 hoursnet/mlx5: reorganize flow tables with hash listBing Zhao
In the current flow tables organization, arrays are used. This is fast for searching, creating related object that will be used in flow creation. But it introduces some limitation to the table index. Then we can reorganize the flow tables information with hash list. When using hash list, there is no need to maintain three arrays for NIC TX, RX and FDB tables object information. This attribute could be used together with the table ID to generate a 64-bits key that is unique for the hash list insertion, lookup and deletion. Signed-off-by: Bing Zhao <bingz@mellanox.com> Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
35 hoursnet/mlx5: control transmit doorbell register mappingViacheslav Ovsiienko
The rdma core library can map doorbell register in two ways, depending on the environment variable "MLX5_SHUT_UP_BF": - as regular cached memory, the variable is either missing or set to zero. This type of mapping may cause the significant doorbell register writing latency and requires explicit memory write barrier to mitigate this issue and prevent write combining. - as non-cached memory, the variable is present and set to not "0" value. This type of mapping may cause performance impact under heavy loading conditions but the explicit write memory barrier is not required and it may improve core performance. The new devarg is introduced "tx_db_nc", if this parameter is set to zero, the doorbell register is forced to be mapped to cached memory and requires explicit memory barrier after writing to. If "tx_db_nc" is set to non-zero value the doorbell will be mapped as non-cached memory, not requiring the memory barrier. If "tx_db_nc" is missing the behaviour will be defined by presence of "MLX5_SHUT_UP_BF" in environment. If variable is missed the default value zero will be set for ARM64 hosts and one for others. In run time the code checks the mapping type and provides the memory barrier after writing to tx doorbell register if it is needed. The mapping type is extracted directly from the uar_mmap_offset field in the queue properties. Fixes: 18a1c20044c0 ("net/mlx5: implement Tx burst template") Cc: stable@dpdk.org Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/failsafe: support xstatsStephen Hemminger
Add support for extended statistics in failsafe driver. Reports detailed statistics for each sub device. Example: testpmd> show port xstats 1 rx_good_packets: 0 tx_good_packets: 0 rx_good_bytes: 0 tx_good_bytes: 0 rx_missed_errors: 0 rx_errors: 0 tx_errors: 0 rx_mbuf_allocation_errors: 0 rx_q0packets: 0 rx_q0bytes: 0 rx_q0errors: 0 tx_q0packets: 0 tx_q0bytes: 0 rx_sub0_good_packets: 0 tx_sub0_good_packets: 0 ... rx_sub1_good_packets: 0 tx_sub1_good_packets: 0 rx_sub1_good_bytes: 0 Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
35 hoursnet/ice: free HW tables when device closedQi Zhang
Free the HW tables during dev_close. Otherwise there will be two issues: 1. Memory won't be freed if the device is detached. 2. Driver can't be initialized correctly after device reset. Fixes: a4c8c48fe3f4 ("net/ice: load OS default package") Cc: stable@dpdk.org Signed-off-by: Qi Zhang <qi.z.zhang@intel.com> Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
35 hoursnet/ice: remove VLAN TPID setTao Zhu
Set vlan tpid is not supported by ice hardware. Delete driver code that doesn't work. Signed-off-by: Tao Zhu <taox.zhu@intel.com> Acked-by: Qiming Yang <qiming.yang@intel.com> Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
35 hoursnet/ice: add error cases for wrong packagesSimei Su
If the switch/fdir/hash module can't resolve a correct parser, it should return error. Fixes: 47d460d63233 ("net/ice: rework switch filter") Fixes: efc16c621415 ("net/ice: support flow director GTPU tunnel") Fixes: 5ad3db8d4bdd ("net/ice: enable advanced RSS") Signed-off-by: Simei Su <simei.su@intel.com> Acked-by: Qi Zhang <qi.z.zhang@intel.com> Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
35 hoursnet/mlx5: fix asserts for multi-segment Tx offload misconfigViacheslav Ovsiienko
To support multi-segment packet sending the Tx datapath DEV_TX_OFFLOAD_MULTI_SEGS offload must be configured. In debug configuration the asserts arise if the multi-segment packet is being sent and no DEV_TX_OFFLOAD_MULTI_SEGS is set. It was found that in some execution paths asserts were missed, this patch adds ones. Fixes: 18a1c20044c0 ("net/mlx5: implement Tx burst template") Cc: stable@dpdk.org Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: clean meter resourcesSuanming Mou
When the port is closed or program exits ungraceful, the meter rulers should be flushed after the flow destroyed. Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: share tag between meter and metadataSuanming Mou
In the meter flow split, metadata flow will be as the sub flow of meter suffix flow. In meter suffix flow, there is already a unique id tag exist as for the meter prefix and suffix flow match. Make metadata feature and meter both share the unique id tag for match. Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: split meter flowSuanming Mou
The flow with meter action will be divided to three sub flows, the prefix flow, the meter flow and the suffix flow. For these three sub flows, as the prefix flow and meter flow have the meter color match. In order the make the packet from prefix flow to match with the suffix flow, an extra metadata register is allocated. The prefix flow will add a unique id to the register and the suffix flow matches on that unique id. As RSS will also divided the flow to several sub flows, flow with meter will be divided as the sub flow of the RSS sub flows if have as below: Original flow -> RSS sub flow 1 -> Meter sub flow 1 (Contain three sub flows.) RSS sub flow 2 -> Meter sub flow 2 (Contain three sub flows.) ...... RSS sub flow n -> Meter sub flow n (Contain three sub flows.) The metadata feature flow will be split as the sub flow of the meter suffix flow. Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: support meter flow actionSuanming Mou
Add meter flow action support in flow validate and translate. Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: add meter attach and detachSuanming Mou
Add the meter attach and detach for the flow create. When create the flow with meter, first try to find any created meter action matches the flow meter id. If the meter action is already created, just attach to it and increase the ref_cnt. If not, create one. For the dettach, decrease the ref_cnt, destroy the meter action while the ref_cnt decreased to zero. Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: add meter statistics read and updateSuanming Mou
This commit add the meter statistics read and update to check the meter statistics. New internal functions in rte_mtr_ops callback: 1. stats_update() 2. stats_read() Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: add count action to meterSuanming Mou
Add count action to meter for metering packet statistics. All the packets be colored and dropped will be recorded. Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: expose flow counters managementSuanming Mou
Expose the flow counter management mechanism for other components to use. Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: support meter profile updateSuanming Mou
This commit add the meter profile update support. New internal function in rte_mtr_ops callback: 1. meter_profile_update() Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: support meter modification operationsSuanming Mou
This commit add meter enable and disable supoort. New internal functions in rte_mtr_ops callback: 1. meter_enable() 2. meter_disable() The meter_enable() enables the meter action and the meter_disable() disables the meter action. Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: add meter action creation to the glueSuanming Mou
This commit add the meter action creation to the glue code. Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: support basic meter operationsSuanming Mou
This commit add the basic meter operations for meter create and destroy. New internal functions in rte_mtr_ops callback: 1. create() 2. destroy() The create() callback will create the corresponding flow rules on the meter table. The destroy() callback destroys the flow rules on the meter table. Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: add policer rules operationsSuanming Mou
This commit create the color rules on the meter table for the packets. As the prefix flow with meter action colors the packets, the packets are transferred to the meter table with meter color match flows. Here we create the flow rules with green yellow red actions on the meter table. Packets match the color will be processed by the related color flow rule. Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: prepare meter flow tablesSuanming Mou
This commit prepare the meter table and suffix table. A flow with meter will be split to three flows. The three flows are created on differnet tables. The packets transfer between the flows on the tables as below: Prefix flow -> Meter flow -> Suffix flow Prefix flow does the user defined match and the meter action. The meter action colors the packet and set its destination to meter table to be processed by the meter flow. The meter flow judges if the packet can be passed or not. If packet can be passed, it will be transferred to the suffix table. The suffix flow on the suffix table will apply the left user defined actions to the packet. The ingress egress and transfer all have the independent meter and suffix tables. Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: validate meter profileSuanming Mou
The add meter profile should be validated if it is valid or has been add to the list. Invalid and exist profile should not be add to the list. Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: support meter profile operationsSuanming Mou
This commit add the support of meter profile add and delete operations. New internal functions in rte_mtr_ops callback: 1. meter_profile_add() 2. meter_profile_delete() Only RTE_MTR_SRTCM_RFC2697 algorithm is supported and can be added. To add other algorithm will report an error. Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: allocate flow meter registersSuanming Mou
Meter need the metadata REG_C to have the color match between the prefix flow and the meter flow. As the user define or metadata feature will both use the REG_C in the suffix flow, the color match register meter uses will not impact the register use in the later sub flow. Another case is that tag is add before meter flow. In this case, meter should not touch the register the tag action is using. To avoid that case, meter should reserve the REG_C's used by user defined MLX5_APP_TAG. Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: fill meter capabilities using DevXSuanming Mou
This commit add the support of fill and get the meter capabilities from DevX. Support items: 1. The srTCM color bind mode. 2. Meter share with multiple flows. 3. Action drop. The color aware mode and multiple meter chaining in a flow are not supported. New internal function in rte_mtr_ops callback: 1. capabilities_get() Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: add meter operation callbackSuanming Mou
Add the new mlx5_flow_meter.c file for metering support. Signed-off-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: fix condition to create default ruleDekel Peled
Previous patch added creation of a default flow rule on port start. Rule is created under the condition that device is in eswitch mode, and is not a VF, to make sure rule is created only once. In Bluefield, where PF representor is used, this condition is not sufficient. Rule is created twice, causing loss of traffic. This patch updates this condition, adding check that device is also not a representor. Fixes: b67b4ecbde22 ("net/mlx5: skip table zero to improve insertion rate") Signed-off-by: Dekel Peled <dekelp@mellanox.com> Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
35 hoursnet/mlx5: allow jump to group lower than currentDekel Peled
In current implementation, jump action is allowed only if target group is higher than the current flow group, This patch updates function flow_dv_validate_action_jump() to allow jump action if target group is higher or lower than the current flow group. Target group equal to current flow group is still rejected. Signed-off-by: Dekel Peled <dekelp@mellanox.com> Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
35 hoursnet/mlx5: add metadata register copy tableViacheslav Ovsiienko
While reg_c[meta] can be copied to reg_b simply by modify-header action (it is supported by hardware), it is not possible to copy reg_c[mark] to the STE flow_tag as flow_tag is not a metadata register and this is not supported by hardware. Instead, it should be manually set by a flow per each unique MARK ID. For this purpose, there should be a dedicated flow table - RX_CP_TBL and all the Rx flow should pass by the table to properly copy values from the register to flow tag field. And for each MARK action, a copy flow should be added to RX_CP_TBL according to the MARK ID like: (if reg_c[mark] == mark_id), flow_tag := mark_id / reg_b := reg_c[meta] / jump to RX_ACT_TBL For SET_META action, there can be only one default flow like: reg_b := reg_c[meta] / jump to RX_ACT_TBL Signed-off-by: Yongseok Koh <yskoh@mellanox.com> Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: split Rx flows to provide metadata copyViacheslav Ovsiienko
Values set by MARK and SET_META actions should be carried over to the VF representor in case of flow miss on Tx path. However, as not all metadata registers are preserved across the different domains (NIC Rx/Tx and E-Switch FDB), as a workaround, those values should be carried by reg_c's which are preserved across domains and copied to STE flow_tag (MARK) and reg_b (META) fields in the last stage of flow steering, in order to scatter those values to flow_tag and flow_table_metadata of CQE. While reg_c[meta] can be copied to reg_b simply by modify-header action (it is supported by hardware), it is not possible to copy reg_c[mark] to the STE flow_tag as flow_tag is not a metadata register and this is not supported by hardware. Instead, it should be manually set by a flow per MARK ID. For this purpose, there should be a dedicated flow table - RX_CP_TBL and all the Rx flow should pass by the table to properly copy values. As the last action of Rx flow steering must be a terminal action such as QUEUE, RSS or DROP, if a user flow has Q/RSS action, the flow must be split in order to pass by the RX_CP_TBL. And the remained Q/RSS action will be performed by another dedicated action table - RX_ACT_TBL. For example, for an ingress flow: pattern, actions_having_QRSS it must be split into two flows. The first one is, pattern, actions_except_QRSS / copy (reg_c[2] := flow_id) / jump to RX_CP_TBL and the second one in RX_ACT_TBL. (if reg_c[2] == flow_id), action_QRSS where flow_id is uniquely allocated and managed identifier. This patch implements the Rx flow splitting and build the RX_ACT_TBL. Also, per each egress flow on NIC Tx, a copy action (reg_c[]= reg_a) should be added in order to transfer metadata from WQE. Signed-off-by: Yongseok Koh <yskoh@mellanox.com> Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: introduce flow splitters chainViacheslav Ovsiienko
The mlx5 hardware has some limitations and flow might require to be split into multiple internal subflows. For example this is needed to provide the meter object sharing between multiple flows or to provide metadata register copying before final queue/rss action. The multiple features might require several level of splitting. For example, hairpin feature splits the original flow into two ones - rx and tx parts. Then RSS feature should split rx part into multiple subflows with extended item sets. Then, metering feature might require splitting each RSS subflow into meter jump chain, and then metadata extensive support might require the final subflows splitting. So, we have to organize the chain of splitting subroutines to abstract each level of splitting. Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: add metadata support to Rx datapathViacheslav Ovsiienko
This patch moves metadata from completion descriptor to appropriate dynamic mbuf field. Signed-off-by: Yongseok Koh <yskoh@mellanox.com> Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: extend flow metadata supportViacheslav Ovsiienko
META item is supported on both Rx and Tx. 'transfer' attribute is also supported. SET_META action is also added. Due to restriction on reg_c[meta], various bit width might be available. If devarg parameter dv_xmeta_en=1, the META uses metadata register reg_c[0], which may be required for internal kernel or firmware needs. In this case PMD queries kernel about available fields in reg_c[0] and restricts the register usage accordingly. If devarg parameter dv_xmeta_en=2, the META feature uses reg_c[1], there should be no limitations on the data width. However, extensive MEAT feature is currently disabled until register copy on loopback is supported by forthcoming patches. Signed-off-by: Yongseok Koh <yskoh@mellanox.com> Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: extend flow mark supportViacheslav Ovsiienko
Flow MARK item is newly supported along with MARK action. MARK action and item are supported on both Rx and Tx. It works on the metadata reg_c[] only if extensive flow metadata register is supported. Without the support, MARK action behaves same as before - valid only on Rx and no MARK item is valid. FLAG action is also modified accordingly. FLAG action is supported on both Rx and Tx via reg_c[] if extensive flow metadata register is supported. However, the new MARK/FLAG item and action are currently disabled until register copy on loopback is supported by forthcoming patches. The actual index of engaged metadata reg_c[] register to support FLAG/MARK actions depends on dv_xmeta_en devarg value. For extensive metadata mode 1 the reg_c[1] is used and transitive MARK data width is 24. For extensive metadata mode 2 the reg_c[0] is used and transitive MARK data width might be restricted to 0 or 16 bits, depending on kernel usage of reg_c[0]. The actual supported width can be discovered by series of trials with rte_flow_validate(). Signed-off-by: Yongseok Koh <yskoh@mellanox.com> Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: support flow tagViacheslav Ovsiienko
Add support of new rte_flow item and action - TAG and SET_TAG. TAG is a transient value which can be kept during flow matching. This is supported through device metadata register reg_c[]. Although there are 8 registers are available on the current mlx5 device, some of them can be reserved for firmware or kernel purposes. The availability should be queried by iterative trial-and-error mlx5_flow_discover_mreg_c() routine. Signed-off-by: Yongseok Koh <yskoh@mellanox.com> Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: update metadata register ID queryViacheslav Ovsiienko
The NIC might support up to 8 extensive metadata registers. These registers are supposed to be used by multiple features. There is register id query routine to allow determine which register is actually used by specified feature. Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: check maximum modify actions numberViacheslav Ovsiienko
If the extensive metadata registers are supported, it can be regarded inclusively that the extensive metadata support is possible. E.g. metadata register copy action, supporting 16 modify header actions, reserving register across different steering domain (FDB and NIC) and so on. This patch handles the maximal amount of header modify actions depending on discovered metadata registers support. Signed-off-by: Yongseok Koh <yskoh@mellanox.com> Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: adjust shared register according to maskViacheslav Ovsiienko
The metadata register reg_c[0] might be used by kernel or firmware for their internal purposes. The actual used mask can be queried from the kernel. The remaining bits can be used by PMD to provide META or MARK feature. The code queries the mask of reg_c[0] and adjust the resource usage dynamically. Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: add devarg for extensive metadata supportViacheslav Ovsiienko
The PMD parameter dv_xmeta_en is added to control extensive metadata support. A nonzero value enables extensive flow metadata support if device is capable and driver supports it. This can enable extensive support of MARK and META item of rte_flow. The newly introduced SET_TAG and SET_META actions do not depend on dv_xmeta_en parameter, because there is no compatibility issue for new entities. The dv_xmeta_en is disabled by default. There are some possible configurations, depending on parameter value: - 0, this is default value, defines the legacy mode, the MARK and META related actions and items operate only within NIC Tx and NIC Rx steering domains, no MARK and META information crosses the domain boundaries. The MARK item is 24 bits wide, the META item is 32 bits wide. - 1, this engages extensive metadata mode, the MARK and META related actions and items operate within all supported steering domains, including FDB, MARK and META information may cross the domain boundaries. The ``MARK`` item is 24 bits wide, the META item width depends on kernel and firmware configurations and might be 0, 16 or 32 bits. Within NIC Tx domain META data width is 32 bits for compatibility, the actual width of data transferred to the FDB domain depends on kernel configuration and may be vary. The actual supported width can be retrieved in runtime by series of rte_flow_validate() trials. - 2, this engages extensive metadata mode, the MARK and META related actions and items operate within all supported steering domains, including FDB, MARK and META information may cross the domain boundaries. The META item is 32 bits wide, the MARK item width depends on kernel and firmware configurations and might be 0, 16 or 24 bits. The actual supported width can be retrieved in runtime by series of rte_flow_validate() trials. If there is no E-Switch configuration the ``dv_xmeta_en`` parameter is ignored and the device is configured to operate in legacy mode (0). Signed-off-by: Yongseok Koh <yskoh@mellanox.com> Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: check metadata registers availabilityViacheslav Ovsiienko
The metadata registers reg_c provide support for TAG and SET_TAG features. Although there are 8 registers are available on the current mlx5 devices, some of them can be reserved. The availability should be queried by iterative trial-and-error implemented by mlx5_flow_discover_mreg_c() routine. If reg_c is available, it can be regarded inclusively that the extensive metadata support is possible. E.g. metadata register copy action, supporting 16 modify header actions (instead of 8 by default) preserving register across different domains (FDB and NIC) and so on. Signed-off-by: Yongseok Koh <yskoh@mellanox.com> Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
35 hoursnet/mlx5: rename structure and functionViacheslav Ovsiienko
There are some renaming: - in the DV flow engine overall: flow_d_* -> flow_dv_* - in flow_dv_translate(): res -> mhdr_res Signed-off-by: Yongseok Koh <yskoh@mellanox.com> Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>