From 0a66e976430402d0064687853fddc33088faa98a Mon Sep 17 00:00:00 2001 From: Lukasz Plachno Date: Wed, 3 Apr 2024 12:24:01 +0200 Subject: [PATCH 01/25] ice: Remove unnecessary argument from ice_fdir_comp_rules() Passing v6 argument is unnecessary as flow_type is still analyzed inside the function. Reviewed-by: Przemek Kitszel Signed-off-by: Lukasz Plachno Reviewed-by: Simon Horman Tested-by: Pucha Himasekhar Reddy (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_fdir.c | 85 +++++++++++------------ 1 file changed, 39 insertions(+), 46 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_fdir.c b/drivers/net/ethernet/intel/ice/ice_fdir.c index 5840c3e04a5b..1f7b26f38818 100644 --- a/drivers/net/ethernet/intel/ice/ice_fdir.c +++ b/drivers/net/ethernet/intel/ice/ice_fdir.c @@ -1189,52 +1189,54 @@ static int ice_cmp_ipv6_addr(__be32 *a, __be32 *b) * ice_fdir_comp_rules - compare 2 filters * @a: a Flow Director filter data structure * @b: a Flow Director filter data structure - * @v6: bool true if v6 filter * * Returns true if the filters match */ static bool -ice_fdir_comp_rules(struct ice_fdir_fltr *a, struct ice_fdir_fltr *b, bool v6) +ice_fdir_comp_rules(struct ice_fdir_fltr *a, struct ice_fdir_fltr *b) { enum ice_fltr_ptype flow_type = a->flow_type; /* The calling function already checks that the two filters have the * same flow_type. */ - if (!v6) { - if (flow_type == ICE_FLTR_PTYPE_NONF_IPV4_TCP || - flow_type == ICE_FLTR_PTYPE_NONF_IPV4_UDP || - flow_type == ICE_FLTR_PTYPE_NONF_IPV4_SCTP) { - if (a->ip.v4.dst_ip == b->ip.v4.dst_ip && - a->ip.v4.src_ip == b->ip.v4.src_ip && - a->ip.v4.dst_port == b->ip.v4.dst_port && - a->ip.v4.src_port == b->ip.v4.src_port) - return true; - } else if (flow_type == ICE_FLTR_PTYPE_NONF_IPV4_OTHER) { - if (a->ip.v4.dst_ip == b->ip.v4.dst_ip && - a->ip.v4.src_ip == b->ip.v4.src_ip && - a->ip.v4.l4_header == b->ip.v4.l4_header && - a->ip.v4.proto == b->ip.v4.proto && - a->ip.v4.ip_ver == b->ip.v4.ip_ver && - a->ip.v4.tos == b->ip.v4.tos) - return true; - } - } else { - if (flow_type == ICE_FLTR_PTYPE_NONF_IPV6_UDP || - flow_type == ICE_FLTR_PTYPE_NONF_IPV6_TCP || - flow_type == ICE_FLTR_PTYPE_NONF_IPV6_SCTP) { - if (a->ip.v6.dst_port == b->ip.v6.dst_port && - a->ip.v6.src_port == b->ip.v6.src_port && - !ice_cmp_ipv6_addr(a->ip.v6.dst_ip, - b->ip.v6.dst_ip) && - !ice_cmp_ipv6_addr(a->ip.v6.src_ip, - b->ip.v6.src_ip)) - return true; - } else if (flow_type == ICE_FLTR_PTYPE_NONF_IPV6_OTHER) { - if (a->ip.v6.dst_port == b->ip.v6.dst_port && - a->ip.v6.src_port == b->ip.v6.src_port) - return true; - } + switch (flow_type) { + case ICE_FLTR_PTYPE_NONF_IPV4_TCP: + case ICE_FLTR_PTYPE_NONF_IPV4_UDP: + case ICE_FLTR_PTYPE_NONF_IPV4_SCTP: + if (a->ip.v4.dst_ip == b->ip.v4.dst_ip && + a->ip.v4.src_ip == b->ip.v4.src_ip && + a->ip.v4.dst_port == b->ip.v4.dst_port && + a->ip.v4.src_port == b->ip.v4.src_port) + return true; + break; + case ICE_FLTR_PTYPE_NONF_IPV4_OTHER: + if (a->ip.v4.dst_ip == b->ip.v4.dst_ip && + a->ip.v4.src_ip == b->ip.v4.src_ip && + a->ip.v4.l4_header == b->ip.v4.l4_header && + a->ip.v4.proto == b->ip.v4.proto && + a->ip.v4.ip_ver == b->ip.v4.ip_ver && + a->ip.v4.tos == b->ip.v4.tos) + return true; + break; + case ICE_FLTR_PTYPE_NONF_IPV6_UDP: + case ICE_FLTR_PTYPE_NONF_IPV6_TCP: + case ICE_FLTR_PTYPE_NONF_IPV6_SCTP: + if (a->ip.v6.dst_port == b->ip.v6.dst_port && + a->ip.v6.src_port == b->ip.v6.src_port && + !ice_cmp_ipv6_addr(a->ip.v6.dst_ip, + b->ip.v6.dst_ip) && + !ice_cmp_ipv6_addr(a->ip.v6.src_ip, + b->ip.v6.src_ip)) + return true; + break; + case ICE_FLTR_PTYPE_NONF_IPV6_OTHER: + if (a->ip.v6.dst_port == b->ip.v6.dst_port && + a->ip.v6.src_port == b->ip.v6.src_port) + return true; + break; + default: + break; } return false; @@ -1253,19 +1255,10 @@ bool ice_fdir_is_dup_fltr(struct ice_hw *hw, struct ice_fdir_fltr *input) bool ret = false; list_for_each_entry(rule, &hw->fdir_list_head, fltr_node) { - enum ice_fltr_ptype flow_type; - if (rule->flow_type != input->flow_type) continue; - flow_type = input->flow_type; - if (flow_type == ICE_FLTR_PTYPE_NONF_IPV4_TCP || - flow_type == ICE_FLTR_PTYPE_NONF_IPV4_UDP || - flow_type == ICE_FLTR_PTYPE_NONF_IPV4_SCTP || - flow_type == ICE_FLTR_PTYPE_NONF_IPV4_OTHER) - ret = ice_fdir_comp_rules(rule, input, false); - else - ret = ice_fdir_comp_rules(rule, input, true); + ret = ice_fdir_comp_rules(rule, input); if (ret) { if (rule->fltr_id == input->fltr_id && rule->q_index != input->q_index) From ae67389c53928418975b46d13b51fde4e40ee15f Mon Sep 17 00:00:00 2001 From: Jakub Buchocki Date: Wed, 3 Apr 2024 12:24:02 +0200 Subject: [PATCH 02/25] ice: Implement 'flow-type ether' rules Add support for 'flow-type ether' Flow Director rules via ethtool. Create packet segment info for filter configuration based on ethtool command parameters. Reuse infrastructure already created for ipv4 and ipv6 flows to convert packet segment into extraction sequence, which is later used to program the filter inside Flow Director block of the Rx pipeline. Rules not containing masks are processed by the Flow Director, and support the following set of input parameters in all combinations: src, dst, proto, vlan-etype, vlan, action. It is possible to specify address mask in ethtool parameters but only 00:00:00:00:00 and FF:FF:FF:FF:FF are valid. The same applies to proto, vlan-etype and vlan masks: only 0x0000 and 0xffff masks are valid. Testing: (DUT) iperf3 -s (DUT) ethtool -U ens785f0np0 flow-type ether dst \ action 10 (DUT) watch 'ethtool -S ens785f0np0 | grep rx_queue' (LP) iperf3 -c ${DUT_IP} Counters increase only for: 'rx_queue_10_packets' 'rx_queue_10_bytes' Signed-off-by: Jakub Buchocki Co-developed-by: Mateusz Pacuszka Signed-off-by: Mateusz Pacuszka Reviewed-by: Przemek Kitszel Signed-off-by: Lukasz Plachno Reviewed-by: Jacob Keller Reviewed-by: Simon Horman Tested-by: Pucha Himasekhar Reddy (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- .../net/ethernet/intel/ice/ice_ethtool_fdir.c | 140 +++++++++++++++++- drivers/net/ethernet/intel/ice/ice_fdir.c | 26 ++++ drivers/net/ethernet/intel/ice/ice_fdir.h | 5 + drivers/net/ethernet/intel/ice/ice_type.h | 1 + 4 files changed, 171 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool_fdir.c b/drivers/net/ethernet/intel/ice/ice_ethtool_fdir.c index 9a1a04f5f146..e3cab8e98f52 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool_fdir.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool_fdir.c @@ -41,6 +41,8 @@ static struct in6_addr zero_ipv6_addr_mask = { static int ice_fltr_to_ethtool_flow(enum ice_fltr_ptype flow) { switch (flow) { + case ICE_FLTR_PTYPE_NONF_ETH: + return ETHER_FLOW; case ICE_FLTR_PTYPE_NONF_IPV4_TCP: return TCP_V4_FLOW; case ICE_FLTR_PTYPE_NONF_IPV4_UDP: @@ -72,6 +74,8 @@ static int ice_fltr_to_ethtool_flow(enum ice_fltr_ptype flow) static enum ice_fltr_ptype ice_ethtool_flow_to_fltr(int eth) { switch (eth) { + case ETHER_FLOW: + return ICE_FLTR_PTYPE_NONF_ETH; case TCP_V4_FLOW: return ICE_FLTR_PTYPE_NONF_IPV4_TCP; case UDP_V4_FLOW: @@ -137,6 +141,10 @@ int ice_get_ethtool_fdir_entry(struct ice_hw *hw, struct ethtool_rxnfc *cmd) memset(&fsp->m_ext, 0, sizeof(fsp->m_ext)); switch (fsp->flow_type) { + case ETHER_FLOW: + fsp->h_u.ether_spec = rule->eth; + fsp->m_u.ether_spec = rule->eth_mask; + break; case IPV4_USER_FLOW: fsp->h_u.usr_ip4_spec.ip_ver = ETH_RX_NFC_IP4; fsp->h_u.usr_ip4_spec.proto = 0; @@ -1193,6 +1201,122 @@ ice_set_fdir_ip6_usr_seg(struct ice_flow_seg_info *seg, return 0; } +/** + * ice_fdir_vlan_valid - validate VLAN data for Flow Director rule + * @dev: network interface device structure + * @fsp: pointer to ethtool Rx flow specification + * + * Return: true if vlan data is valid, false otherwise + */ +static bool ice_fdir_vlan_valid(struct device *dev, + struct ethtool_rx_flow_spec *fsp) +{ + if (fsp->m_ext.vlan_etype && !eth_type_vlan(fsp->h_ext.vlan_etype)) + return false; + + if (fsp->m_ext.vlan_tci && ntohs(fsp->h_ext.vlan_tci) >= VLAN_N_VID) + return false; + + /* proto and vlan must have vlan-etype defined */ + if (fsp->m_u.ether_spec.h_proto && fsp->m_ext.vlan_tci && + !fsp->m_ext.vlan_etype) { + dev_warn(dev, "Filter with proto and vlan require also vlan-etype"); + return false; + } + + return true; +} + +/** + * ice_set_ether_flow_seg - set address and protocol segments for ether flow + * @dev: network interface device structure + * @seg: flow segment for programming + * @eth_spec: mask data from ethtool + * + * Return: 0 on success and errno in case of error. + */ +static int ice_set_ether_flow_seg(struct device *dev, + struct ice_flow_seg_info *seg, + struct ethhdr *eth_spec) +{ + ICE_FLOW_SET_HDRS(seg, ICE_FLOW_SEG_HDR_ETH); + + /* empty rules are not valid */ + if (is_zero_ether_addr(eth_spec->h_source) && + is_zero_ether_addr(eth_spec->h_dest) && + !eth_spec->h_proto) + return -EINVAL; + + /* Ethertype */ + if (eth_spec->h_proto == htons(0xFFFF)) { + ice_flow_set_fld(seg, ICE_FLOW_FIELD_IDX_ETH_TYPE, + ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, false); + } else if (eth_spec->h_proto) { + dev_warn(dev, "Only 0x0000 or 0xffff proto mask is allowed for flow-type ether"); + return -EOPNOTSUPP; + } + + /* Source MAC address */ + if (is_broadcast_ether_addr(eth_spec->h_source)) + ice_flow_set_fld(seg, ICE_FLOW_FIELD_IDX_ETH_SA, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, false); + else if (!is_zero_ether_addr(eth_spec->h_source)) + goto err_mask; + + /* Destination MAC address */ + if (is_broadcast_ether_addr(eth_spec->h_dest)) + ice_flow_set_fld(seg, ICE_FLOW_FIELD_IDX_ETH_DA, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, false); + else if (!is_zero_ether_addr(eth_spec->h_dest)) + goto err_mask; + + return 0; + +err_mask: + dev_warn(dev, "Only 00:00:00:00:00:00 or ff:ff:ff:ff:ff:ff MAC address mask is allowed for flow-type ether"); + return -EOPNOTSUPP; +} + +/** + * ice_set_fdir_vlan_seg - set vlan segments for ether flow + * @seg: flow segment for programming + * @ext_masks: masks for additional RX flow fields + * + * Return: 0 on success and errno in case of error. + */ +static int +ice_set_fdir_vlan_seg(struct ice_flow_seg_info *seg, + struct ethtool_flow_ext *ext_masks) +{ + ICE_FLOW_SET_HDRS(seg, ICE_FLOW_SEG_HDR_VLAN); + + if (ext_masks->vlan_etype) { + if (ext_masks->vlan_etype != htons(0xFFFF)) + return -EOPNOTSUPP; + + ice_flow_set_fld(seg, ICE_FLOW_FIELD_IDX_S_VLAN, + ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, false); + } + + if (ext_masks->vlan_tci) { + if (ext_masks->vlan_tci != htons(0xFFFF)) + return -EOPNOTSUPP; + + ice_flow_set_fld(seg, ICE_FLOW_FIELD_IDX_C_VLAN, + ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, false); + } + + return 0; +} + /** * ice_cfg_fdir_xtrct_seq - Configure extraction sequence for the given filter * @pf: PF structure @@ -1209,7 +1333,7 @@ ice_cfg_fdir_xtrct_seq(struct ice_pf *pf, struct ethtool_rx_flow_spec *fsp, struct device *dev = ice_pf_to_dev(pf); enum ice_fltr_ptype fltr_idx; struct ice_hw *hw = &pf->hw; - bool perfect_filter; + bool perfect_filter = false; int ret; seg = devm_kzalloc(dev, sizeof(*seg), GFP_KERNEL); @@ -1262,6 +1386,16 @@ ice_cfg_fdir_xtrct_seq(struct ice_pf *pf, struct ethtool_rx_flow_spec *fsp, ret = ice_set_fdir_ip6_usr_seg(seg, &fsp->m_u.usr_ip6_spec, &perfect_filter); break; + case ETHER_FLOW: + ret = ice_set_ether_flow_seg(dev, seg, &fsp->m_u.ether_spec); + if (!ret && (fsp->m_ext.vlan_etype || fsp->m_ext.vlan_tci)) { + if (!ice_fdir_vlan_valid(dev, fsp)) { + ret = -EINVAL; + break; + } + ret = ice_set_fdir_vlan_seg(seg, &fsp->m_ext); + } + break; default: ret = -EINVAL; } @@ -1823,6 +1957,10 @@ ice_set_fdir_input_set(struct ice_vsi *vsi, struct ethtool_rx_flow_spec *fsp, input->mask.v6.tc = fsp->m_u.usr_ip6_spec.tclass; input->mask.v6.proto = fsp->m_u.usr_ip6_spec.l4_proto; break; + case ETHER_FLOW: + input->eth = fsp->h_u.ether_spec; + input->eth_mask = fsp->m_u.ether_spec; + break; default: /* not doing un-parsed flow types */ return -EINVAL; diff --git a/drivers/net/ethernet/intel/ice/ice_fdir.c b/drivers/net/ethernet/intel/ice/ice_fdir.c index 1f7b26f38818..26b357c0ae15 100644 --- a/drivers/net/ethernet/intel/ice/ice_fdir.c +++ b/drivers/net/ethernet/intel/ice/ice_fdir.c @@ -4,6 +4,8 @@ #include "ice_common.h" /* These are training packet headers used to program flow director filters. */ +static const u8 ice_fdir_eth_pkt[22]; + static const u8 ice_fdir_tcpv4_pkt[] = { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, 0x45, 0x00, @@ -416,6 +418,11 @@ static const u8 ice_fdir_ip6_tun_pkt[] = { /* Flow Director no-op training packet table */ static const struct ice_fdir_base_pkt ice_fdir_pkt[] = { + { + ICE_FLTR_PTYPE_NONF_ETH, + sizeof(ice_fdir_eth_pkt), ice_fdir_eth_pkt, + sizeof(ice_fdir_eth_pkt), ice_fdir_eth_pkt, + }, { ICE_FLTR_PTYPE_NONF_IPV4_TCP, sizeof(ice_fdir_tcpv4_pkt), ice_fdir_tcpv4_pkt, @@ -914,6 +921,21 @@ ice_fdir_get_gen_prgm_pkt(struct ice_hw *hw, struct ice_fdir_fltr *input, * perspective. The input from user is from Rx filter perspective. */ switch (flow) { + case ICE_FLTR_PTYPE_NONF_ETH: + ice_pkt_insert_mac_addr(loc, input->eth.h_dest); + ice_pkt_insert_mac_addr(loc + ETH_ALEN, input->eth.h_source); + if (input->ext_data.vlan_tag || input->ext_data.vlan_type) { + ice_pkt_insert_u16(loc, ICE_ETH_TYPE_F_OFFSET, + input->ext_data.vlan_type); + ice_pkt_insert_u16(loc, ICE_ETH_VLAN_TCI_OFFSET, + input->ext_data.vlan_tag); + ice_pkt_insert_u16(loc, ICE_ETH_TYPE_VLAN_OFFSET, + input->eth.h_proto); + } else { + ice_pkt_insert_u16(loc, ICE_ETH_TYPE_F_OFFSET, + input->eth.h_proto); + } + break; case ICE_FLTR_PTYPE_NONF_IPV4_TCP: ice_pkt_insert_u32(loc, ICE_IPV4_DST_ADDR_OFFSET, input->ip.v4.src_ip); @@ -1201,6 +1223,10 @@ ice_fdir_comp_rules(struct ice_fdir_fltr *a, struct ice_fdir_fltr *b) * same flow_type. */ switch (flow_type) { + case ICE_FLTR_PTYPE_NONF_ETH: + if (!memcmp(&a->eth, &b->eth, sizeof(a->eth))) + return true; + break; case ICE_FLTR_PTYPE_NONF_IPV4_TCP: case ICE_FLTR_PTYPE_NONF_IPV4_UDP: case ICE_FLTR_PTYPE_NONF_IPV4_SCTP: diff --git a/drivers/net/ethernet/intel/ice/ice_fdir.h b/drivers/net/ethernet/intel/ice/ice_fdir.h index 1b9b84490689..021ecbac7848 100644 --- a/drivers/net/ethernet/intel/ice/ice_fdir.h +++ b/drivers/net/ethernet/intel/ice/ice_fdir.h @@ -8,6 +8,9 @@ #define ICE_FDIR_MAX_RAW_PKT_SIZE (512 + ICE_FDIR_TUN_PKT_OFF) /* macros for offsets into packets for flow director programming */ +#define ICE_ETH_TYPE_F_OFFSET 12 +#define ICE_ETH_VLAN_TCI_OFFSET 14 +#define ICE_ETH_TYPE_VLAN_OFFSET 16 #define ICE_IPV4_SRC_ADDR_OFFSET 26 #define ICE_IPV4_DST_ADDR_OFFSET 30 #define ICE_IPV4_TCP_SRC_PORT_OFFSET 34 @@ -159,6 +162,8 @@ struct ice_fdir_fltr { struct list_head fltr_node; enum ice_fltr_ptype flow_type; + struct ethhdr eth, eth_mask; + union { struct ice_fdir_v4 v4; struct ice_fdir_v6 v6; diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h index 08ec5efdafe6..4049ae40c4fb 100644 --- a/drivers/net/ethernet/intel/ice/ice_type.h +++ b/drivers/net/ethernet/intel/ice/ice_type.h @@ -203,6 +203,7 @@ struct ice_phy_info { enum ice_fltr_ptype { /* NONE - used for undef/error */ ICE_FLTR_PTYPE_NONF_NONE = 0, + ICE_FLTR_PTYPE_NONF_ETH, ICE_FLTR_PTYPE_NONF_IPV4_UDP, ICE_FLTR_PTYPE_NONF_IPV4_TCP, ICE_FLTR_PTYPE_NONF_IPV4_SCTP, From c22f7dacb8202779e60f45976443d979a749ab13 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Fri, 22 Mar 2024 14:44:44 -0700 Subject: [PATCH 03/25] ice: set vf->num_msix in ice_initialize_vf_entry() Commit fe1c5ca2fe76 ("ice: implement num_msix field per VF") updated the driver to allow for per-VF MSI-X configuration. The initial defaults were set in ice_create_vf_entries(). This logic is better placed in ice_initialize_vf_entry(). Indeed, that function already sets vf->num_vf_qs, as well as initializes the allow list via calling ice_vc_set_default_allowlist(). Move this logic into ice_initialize_vf_entry(). This makes the code clear, and ensures that these VF fields will be initialized properly for both SR-IOV VFs and the upcoming Scalable IOV VFs. Signed-off-by: Jacob Keller Reviewed-by: Jesse Brandeburg Tested-by: Rafal Romanowski Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_sriov.c | 5 ----- drivers/net/ethernet/intel/ice/ice_vf_lib.c | 5 ++++- 2 files changed, 4 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.c b/drivers/net/ethernet/intel/ice/ice_sriov.c index 65e1986af777..5e9521876617 100644 --- a/drivers/net/ethernet/intel/ice/ice_sriov.c +++ b/drivers/net/ethernet/intel/ice/ice_sriov.c @@ -831,11 +831,6 @@ static int ice_create_vf_entries(struct ice_pf *pf, u16 num_vfs) pci_dev_get(vfdev); - /* set default number of MSI-X */ - vf->num_msix = pf->vfs.num_msix_per; - vf->num_vf_qs = pf->vfs.num_qps_per; - ice_vc_set_default_allowlist(vf); - hash_add_rcu(vfs->table, &vf->entry, vf_id); } diff --git a/drivers/net/ethernet/intel/ice/ice_vf_lib.c b/drivers/net/ethernet/intel/ice/ice_vf_lib.c index 21d26e19338a..c51e2482cad2 100644 --- a/drivers/net/ethernet/intel/ice/ice_vf_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_vf_lib.c @@ -992,10 +992,13 @@ void ice_initialize_vf_entry(struct ice_vf *vf) /* assign default capabilities */ vf->spoofchk = true; - vf->num_vf_qs = vfs->num_qps_per; ice_vc_set_default_allowlist(vf); ice_virtchnl_set_dflt_ops(vf); + /* set default number of MSI-X */ + vf->num_msix = vfs->num_msix_per; + vf->num_vf_qs = vfs->num_qps_per; + /* ctrl_vsi_idx will be set to a valid value only when iAVF * creates its first fdir rule. */ From b80d01ef9aba6c0824644bbde49b1f4e2955e5d1 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Fri, 22 Mar 2024 14:44:45 -0700 Subject: [PATCH 04/25] ice: store VF relative MSI-X index in q_vector->vf_reg_idx The ice physical function driver needs to configure the association of queues and interrupts on behalf of its virtual functions. This is done over virtchnl by the VF sending messages during its initialization phase. These messages contain a vector_id which the VF wants to associate with a given queue. This ID is relative to the VF space, where 0 indicates the control IRQ for non-queue interrupts. When programming the mapping, the PF driver currently passes this vector_id directly to the low level functions for programming. This works for SR-IOV, because the hardware uses the VF-based indexing for interrupts. This won't work for Scalable IOV, which uses PF-based indexing for programming its VSIs. To handle this, the driver needs to be able to look up the proper index to use for programming. For typical IRQs, this would be the q_vector->reg_idx field. The q_vector->reg_idx can't be set to a VF relative value, because it is used when the PF needs to control the interrupt, such as when triggering a software interrupt on stopping the Tx queue. Thus, introduce a new q_vector->vf_reg_idx which can store the VF relative index for registers which expect this. Use this in ice_cfg_interrupt to look up the VF index from the q_vector. This allows removing the vector ID parameter of ice_cfg_interrupt. Also notice that this function returns an int, but then is cast to the virtchnl error enumeration, virtchnl_status_code. Update the return type to indicate it does not return an integer error code. We can't use normal error codes here because the return values are passed across the virtchnl interface. This will allow the future Scalable IOV VFs to correctly look up the index needed for programming the VF queues without breaking SR-IOV. Signed-off-by: Jacob Keller Tested-by: Rafal Romanowski Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice.h | 3 ++- drivers/net/ethernet/intel/ice/ice_base.c | 3 ++- drivers/net/ethernet/intel/ice/ice_sriov.c | 7 ++++--- drivers/net/ethernet/intel/ice/ice_sriov.h | 5 ++--- drivers/net/ethernet/intel/ice/ice_virtchnl.c | 14 +++++++------- 5 files changed, 17 insertions(+), 15 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index a7e88d797d4c..67a3236ab1fc 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -459,7 +459,7 @@ struct ice_q_vector { struct ice_vsi *vsi; u16 v_idx; /* index in the vsi->q_vector array. */ - u16 reg_idx; + u16 reg_idx; /* PF relative register index */ u8 num_ring_rx; /* total number of Rx rings in vector */ u8 num_ring_tx; /* total number of Tx rings in vector */ u8 wb_on_itr:1; /* if true, WB on ITR is enabled */ @@ -481,6 +481,7 @@ struct ice_q_vector { char name[ICE_INT_NAME_STR_LEN]; u16 total_events; /* net_dim(): number of interrupts processed */ + u16 vf_reg_idx; /* VF relative register index */ struct msi_map irq; } ____cacheline_internodealigned_in_smp; diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c index af0384b37119..687f6cb2b917 100644 --- a/drivers/net/ethernet/intel/ice/ice_base.c +++ b/drivers/net/ethernet/intel/ice/ice_base.c @@ -121,7 +121,7 @@ static int ice_vsi_alloc_q_vector(struct ice_vsi *vsi, u16 v_idx) q_vector->irq.index = -ENOENT; if (vsi->type == ICE_VSI_VF) { - q_vector->reg_idx = ice_calc_vf_reg_idx(vsi->vf, q_vector); + ice_calc_vf_reg_idx(vsi->vf, q_vector); goto out; } else if (vsi->type == ICE_VSI_CTRL && vsi->vf) { struct ice_vsi *ctrl_vsi = ice_get_vf_ctrl_vsi(pf, vsi); @@ -145,6 +145,7 @@ static int ice_vsi_alloc_q_vector(struct ice_vsi *vsi, u16 v_idx) skip_alloc: q_vector->reg_idx = q_vector->irq.index; + q_vector->vf_reg_idx = q_vector->irq.index; /* only set affinity_mask if the CPU is online */ if (cpu_online(v_idx)) diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.c b/drivers/net/ethernet/intel/ice/ice_sriov.c index 5e9521876617..fb2e96db647e 100644 --- a/drivers/net/ethernet/intel/ice/ice_sriov.c +++ b/drivers/net/ethernet/intel/ice/ice_sriov.c @@ -360,13 +360,14 @@ static void ice_ena_vf_mappings(struct ice_vf *vf) * @vf: VF to calculate the register index for * @q_vector: a q_vector associated to the VF */ -int ice_calc_vf_reg_idx(struct ice_vf *vf, struct ice_q_vector *q_vector) +void ice_calc_vf_reg_idx(struct ice_vf *vf, struct ice_q_vector *q_vector) { if (!vf || !q_vector) - return -EINVAL; + return; /* always add one to account for the OICR being the first MSIX */ - return vf->first_vector_idx + q_vector->v_idx + 1; + q_vector->vf_reg_idx = q_vector->v_idx + ICE_NONQ_VECS_VF; + q_vector->reg_idx = vf->first_vector_idx + q_vector->vf_reg_idx; } /** diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.h b/drivers/net/ethernet/intel/ice/ice_sriov.h index 8488df38b586..4ba8fb53aea1 100644 --- a/drivers/net/ethernet/intel/ice/ice_sriov.h +++ b/drivers/net/ethernet/intel/ice/ice_sriov.h @@ -49,7 +49,7 @@ int ice_set_vf_link_state(struct net_device *netdev, int vf_id, int link_state); int ice_set_vf_spoofchk(struct net_device *netdev, int vf_id, bool ena); -int ice_calc_vf_reg_idx(struct ice_vf *vf, struct ice_q_vector *q_vector); +void ice_calc_vf_reg_idx(struct ice_vf *vf, struct ice_q_vector *q_vector); int ice_get_vf_stats(struct net_device *netdev, int vf_id, @@ -130,11 +130,10 @@ ice_set_vf_bw(struct net_device __always_unused *netdev, return -EOPNOTSUPP; } -static inline int +static inline void ice_calc_vf_reg_idx(struct ice_vf __always_unused *vf, struct ice_q_vector __always_unused *q_vector) { - return 0; } static inline int diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl.c b/drivers/net/ethernet/intel/ice/ice_virtchnl.c index 1ff9818b4c84..1c6ce0c4ed4e 100644 --- a/drivers/net/ethernet/intel/ice/ice_virtchnl.c +++ b/drivers/net/ethernet/intel/ice/ice_virtchnl.c @@ -1505,13 +1505,12 @@ error_param: * ice_cfg_interrupt * @vf: pointer to the VF info * @vsi: the VSI being configured - * @vector_id: vector ID * @map: vector map for mapping vectors to queues * @q_vector: structure for interrupt vector * configure the IRQ to queue map */ -static int -ice_cfg_interrupt(struct ice_vf *vf, struct ice_vsi *vsi, u16 vector_id, +static enum virtchnl_status_code +ice_cfg_interrupt(struct ice_vf *vf, struct ice_vsi *vsi, struct virtchnl_vector_map *map, struct ice_q_vector *q_vector) { @@ -1531,7 +1530,8 @@ ice_cfg_interrupt(struct ice_vf *vf, struct ice_vsi *vsi, u16 vector_id, q_vector->num_ring_rx++; q_vector->rx.itr_idx = map->rxitr_idx; vsi->rx_rings[vsi_q_id]->q_vector = q_vector; - ice_cfg_rxq_interrupt(vsi, vsi_q_id, vector_id, + ice_cfg_rxq_interrupt(vsi, vsi_q_id, + q_vector->vf_reg_idx, q_vector->rx.itr_idx); } @@ -1545,7 +1545,8 @@ ice_cfg_interrupt(struct ice_vf *vf, struct ice_vsi *vsi, u16 vector_id, q_vector->num_ring_tx++; q_vector->tx.itr_idx = map->txitr_idx; vsi->tx_rings[vsi_q_id]->q_vector = q_vector; - ice_cfg_txq_interrupt(vsi, vsi_q_id, vector_id, + ice_cfg_txq_interrupt(vsi, vsi_q_id, + q_vector->vf_reg_idx, q_vector->tx.itr_idx); } @@ -1619,8 +1620,7 @@ static int ice_vc_cfg_irq_map_msg(struct ice_vf *vf, u8 *msg) } /* lookout for the invalid queue index */ - v_ret = (enum virtchnl_status_code) - ice_cfg_interrupt(vf, vsi, vector_id, map, q_vector); + v_ret = ice_cfg_interrupt(vf, vsi, map, q_vector); if (v_ret) goto error_param; } From 1580cbcbfe770b0a7fb76735c1a601483335c1c2 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 12 Apr 2024 07:14:31 -0700 Subject: [PATCH 05/25] net: netdevsim: add some fake page pool use Add very basic page pool use so that we can exercise the netlink uAPI in a selftest. Page pool gets created on open, destroyed on close. But we control allocating of a single page thru debugfs. This page may survive past the page pool itself so that we can test orphaned page pools. Link: https://lore.kernel.org/r/20240412141436.828666-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/netdevsim/netdev.c | 93 +++++++++++++++++++++++++++++++ drivers/net/netdevsim/netdevsim.h | 4 ++ 2 files changed, 97 insertions(+) diff --git a/drivers/net/netdevsim/netdev.c b/drivers/net/netdevsim/netdev.c index d7ba447db17c..d127856f8f36 100644 --- a/drivers/net/netdevsim/netdev.c +++ b/drivers/net/netdevsim/netdev.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include #include @@ -299,6 +300,29 @@ static int nsim_get_iflink(const struct net_device *dev) return iflink; } +static int nsim_open(struct net_device *dev) +{ + struct netdevsim *ns = netdev_priv(dev); + struct page_pool_params pp = { 0 }; + + pp.pool_size = 128; + pp.dev = &dev->dev; + pp.dma_dir = DMA_BIDIRECTIONAL; + pp.netdev = dev; + + ns->pp = page_pool_create(&pp); + return PTR_ERR_OR_ZERO(ns->pp); +} + +static int nsim_stop(struct net_device *dev) +{ + struct netdevsim *ns = netdev_priv(dev); + + page_pool_destroy(ns->pp); + + return 0; +} + static const struct net_device_ops nsim_netdev_ops = { .ndo_start_xmit = nsim_start_xmit, .ndo_set_rx_mode = nsim_set_rx_mode, @@ -318,6 +342,8 @@ static const struct net_device_ops nsim_netdev_ops = { .ndo_set_features = nsim_set_features, .ndo_get_iflink = nsim_get_iflink, .ndo_bpf = nsim_bpf, + .ndo_open = nsim_open, + .ndo_stop = nsim_stop, }; static const struct net_device_ops nsim_vf_netdev_ops = { @@ -378,6 +404,60 @@ static const struct netdev_stat_ops nsim_stat_ops = { .get_base_stats = nsim_get_base_stats, }; +static ssize_t +nsim_pp_hold_read(struct file *file, char __user *data, + size_t count, loff_t *ppos) +{ + struct netdevsim *ns = file->private_data; + char buf[3] = "n\n"; + + if (ns->page) + buf[0] = 'y'; + + return simple_read_from_buffer(data, count, ppos, buf, 2); +} + +static ssize_t +nsim_pp_hold_write(struct file *file, const char __user *data, + size_t count, loff_t *ppos) +{ + struct netdevsim *ns = file->private_data; + ssize_t ret; + bool val; + + ret = kstrtobool_from_user(data, count, &val); + if (ret) + return ret; + + rtnl_lock(); + ret = count; + if (val == !!ns->page) + goto exit; + + if (!netif_running(ns->netdev) && val) { + ret = -ENETDOWN; + } else if (val) { + ns->page = page_pool_dev_alloc_pages(ns->pp); + if (!ns->page) + ret = -ENOMEM; + } else { + page_pool_put_full_page(ns->page->pp, ns->page, false); + ns->page = NULL; + } + rtnl_unlock(); + +exit: + return count; +} + +static const struct file_operations nsim_pp_hold_fops = { + .open = simple_open, + .read = nsim_pp_hold_read, + .write = nsim_pp_hold_write, + .llseek = generic_file_llseek, + .owner = THIS_MODULE, +}; + static void nsim_setup(struct net_device *dev) { ether_setup(dev); @@ -485,6 +565,10 @@ nsim_create(struct nsim_dev *nsim_dev, struct nsim_dev_port *nsim_dev_port) err = nsim_init_netdevsim_vf(ns); if (err) goto err_free_netdev; + + ns->pp_dfs = debugfs_create_file("pp_hold", 0600, nsim_dev_port->ddir, + ns, &nsim_pp_hold_fops); + return ns; err_free_netdev: @@ -497,6 +581,8 @@ void nsim_destroy(struct netdevsim *ns) struct net_device *dev = ns->netdev; struct netdevsim *peer; + debugfs_remove(ns->pp_dfs); + rtnl_lock(); peer = rtnl_dereference(ns->peer); if (peer) @@ -511,6 +597,13 @@ void nsim_destroy(struct netdevsim *ns) rtnl_unlock(); if (nsim_dev_port_is_pf(ns->nsim_dev_port)) nsim_exit_netdevsim(ns); + + /* Put this intentionally late to exercise the orphaning path */ + if (ns->page) { + page_pool_put_full_page(ns->page->pp, ns->page, false); + ns->page = NULL; + } + free_netdev(dev); } diff --git a/drivers/net/netdevsim/netdevsim.h b/drivers/net/netdevsim/netdevsim.h index 553c4b9b4f63..7664ab823e29 100644 --- a/drivers/net/netdevsim/netdevsim.h +++ b/drivers/net/netdevsim/netdevsim.h @@ -125,6 +125,10 @@ struct netdevsim { struct debugfs_u32_array dfs_ports[2]; } udp_ports; + struct page_pool *pp; + struct page *page; + struct dentry *pp_dfs; + struct nsim_ethtool ethtool; struct netdevsim __rcu *peer; }; From 72ba6cba0a6e9f06c09187ddbbdc9f80ee93ffb3 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 12 Apr 2024 07:14:32 -0700 Subject: [PATCH 06/25] tools: ynl: don't return None for dumps YNL currently reports None for empty dump: $ cli.py ...netdev.yaml --dump page-pool-get None This doesn't matter for the CLI but when writing YNL based tests having to deal with either list or None is annoying. Limit the None conversion to non-dump ops: $ cli.py ...netdev.yaml --dump page-pool-get [] Reviewed-by: Donald Hunter Link: https://lore.kernel.org/r/20240412141436.828666-3-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/net/ynl/lib/ynl.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py index 0ba5f6fb8747..a67f7b6fef92 100644 --- a/tools/net/ynl/lib/ynl.py +++ b/tools/net/ynl/lib/ynl.py @@ -995,9 +995,11 @@ class YnlFamily(SpecFamily): rsp_msg.update(self._decode_struct(decoded.raw, op.fixed_header)) rsp.append(rsp_msg) + if dump: + return rsp if not rsp: return None - if not dump and len(rsp) == 1: + if len(rsp) == 1: return rsp[0] return rsp From eeb409bde964df1956297b6775ff5f53dd98d556 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 12 Apr 2024 07:14:33 -0700 Subject: [PATCH 07/25] selftests: net: print report check location in python tests Developing Python tests is a bit annoying because when test fails we only print the fail message and no info about which exact check led to it. Print the location (the first line of this example is new): # At /root/ksft-net-drv/./net/nl_netdev.py line 38: # Check failed 0 != 10 not ok 3 nl_netdev.page_pool_check Reviewed-by: Petr Machata Link: https://lore.kernel.org/r/20240412141436.828666-4-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/lib/py/ksft.py | 25 ++++++++++++---------- 1 file changed, 14 insertions(+), 11 deletions(-) diff --git a/tools/testing/selftests/net/lib/py/ksft.py b/tools/testing/selftests/net/lib/py/ksft.py index c7210525981c..b4b0bfff68b0 100644 --- a/tools/testing/selftests/net/lib/py/ksft.py +++ b/tools/testing/selftests/net/lib/py/ksft.py @@ -1,6 +1,7 @@ # SPDX-License-Identifier: GPL-2.0 import builtins +import inspect from .consts import KSFT_MAIN_NAME KSFT_RESULT = None @@ -18,32 +19,34 @@ def ksft_pr(*objs, **kwargs): print("#", *objs, **kwargs) +def _fail(*args): + global KSFT_RESULT + KSFT_RESULT = False + + frame = inspect.stack()[2] + ksft_pr("At " + frame.filename + " line " + str(frame.lineno) + ":") + ksft_pr(*args) + + def ksft_eq(a, b, comment=""): global KSFT_RESULT if a != b: - KSFT_RESULT = False - ksft_pr("Check failed", a, "!=", b, comment) + _fail("Check failed", a, "!=", b, comment) def ksft_true(a, comment=""): - global KSFT_RESULT if not a: - KSFT_RESULT = False - ksft_pr("Check failed", a, "does not eval to True", comment) + _fail("Check failed", a, "does not eval to True", comment) def ksft_in(a, b, comment=""): - global KSFT_RESULT if a not in b: - KSFT_RESULT = False - ksft_pr("Check failed", a, "not in", b, comment) + _fail("Check failed", a, "not in", b, comment) def ksft_ge(a, b, comment=""): - global KSFT_RESULT if a < b: - KSFT_RESULT = False - ksft_pr("Check failed", a, "<", b, comment) + _fail("Check failed", a, "<", b, comment) def ktap_result(ok, cnt=1, case="", comment=""): From 99583b970b9073ea258235e6c794fd515df19c61 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 12 Apr 2024 07:14:34 -0700 Subject: [PATCH 08/25] selftests: net: print full exception on failure Instead of a summary line print the full exception. This makes debugging Python tests much easier. Reviewed-by: Petr Machata Link: https://lore.kernel.org/r/20240412141436.828666-5-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/lib/py/ksft.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/net/lib/py/ksft.py b/tools/testing/selftests/net/lib/py/ksft.py index b4b0bfff68b0..793e4761645e 100644 --- a/tools/testing/selftests/net/lib/py/ksft.py +++ b/tools/testing/selftests/net/lib/py/ksft.py @@ -2,6 +2,7 @@ import builtins import inspect +import traceback from .consts import KSFT_MAIN_NAME KSFT_RESULT = None @@ -85,7 +86,8 @@ def ksft_run(cases, args=()): totals['xfail'] += 1 continue except Exception as e: - for line in str(e).split('\n'): + tb = traceback.format_exc() + for line in tb.strip().split('\n'): ksft_pr("Exception|", line) ktap_result(False, cnt, case) totals['fail'] += 1 From 8554d6e39b6ad967e1debe98550a0c56aaf8c8ea Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 12 Apr 2024 07:14:35 -0700 Subject: [PATCH 09/25] selftests: net: support use of NetdevSimDev under "with" in python Using "with" on an entire driver test env is supported already, but it's also useful to use "with" on an individual nsim. Reviewed-by: Petr Machata Link: https://lore.kernel.org/r/20240412141436.828666-6-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/lib/py/nsim.py | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/net/lib/py/nsim.py b/tools/testing/selftests/net/lib/py/nsim.py index 97457aca7e08..94aa32f59fdb 100644 --- a/tools/testing/selftests/net/lib/py/nsim.py +++ b/tools/testing/selftests/net/lib/py/nsim.py @@ -84,6 +84,17 @@ class NetdevSimDev: for port_index in range(port_count): self.nsims.append(self._make_port(port_index, ifnames[port_index])) + self.removed = False + + def __enter__(self): + return self + + def __exit__(self, ex_type, ex_value, ex_tb): + """ + __exit__ gets called at the end of a "with" block. + """ + self.remove() + def _make_port(self, port_index, ifname): return NetdevSim(self, port_index, ifname, self.ns) @@ -112,7 +123,9 @@ class NetdevSimDev: raise Exception("netdevices did not appear within timeout") def remove(self): - self.ctrl_write("del_device", "%u" % (self.addr, )) + if not self.removed: + self.ctrl_write("del_device", "%u" % (self.addr, )) + self.removed = True def remove_nsim(self, nsim): self.nsims.remove(nsim) From 05fa5c31b9882b175d5e5a8e310f31b1a9b019a3 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 12 Apr 2024 07:14:36 -0700 Subject: [PATCH 10/25] selftests: net: exercise page pool reporting via netlink Add a Python test for the basic ops. # ./net/nl_netdev.py KTAP version 1 1..3 ok 1 nl_netdev.empty_check ok 2 nl_netdev.lo_check ok 3 nl_netdev.page_pool_check # Totals: pass:3 fail:0 xfail:0 xpass:0 skip:0 error:0 Reviewed-by: Petr Machata Link: https://lore.kernel.org/r/20240412141436.828666-7-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/lib/py/ksft.py | 12 ++++ tools/testing/selftests/net/lib/py/nsim.py | 1 + tools/testing/selftests/net/nl_netdev.py | 76 +++++++++++++++++++++- 3 files changed, 87 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/lib/py/ksft.py b/tools/testing/selftests/net/lib/py/ksft.py index 793e4761645e..3769b9197213 100644 --- a/tools/testing/selftests/net/lib/py/ksft.py +++ b/tools/testing/selftests/net/lib/py/ksft.py @@ -2,6 +2,7 @@ import builtins import inspect +import time import traceback from .consts import KSFT_MAIN_NAME @@ -50,6 +51,17 @@ def ksft_ge(a, b, comment=""): _fail("Check failed", a, "<", b, comment) +def ksft_busy_wait(cond, sleep=0.005, deadline=1, comment=""): + end = time.monotonic() + deadline + while True: + if cond(): + return + if time.monotonic() > end: + _fail("Waiting for condition timed out", comment) + return + time.sleep(sleep) + + def ktap_result(ok, cnt=1, case="", comment=""): res = "" if not ok: diff --git a/tools/testing/selftests/net/lib/py/nsim.py b/tools/testing/selftests/net/lib/py/nsim.py index 94aa32f59fdb..06896cdf7c18 100644 --- a/tools/testing/selftests/net/lib/py/nsim.py +++ b/tools/testing/selftests/net/lib/py/nsim.py @@ -28,6 +28,7 @@ class NetdevSim: self.dfs_dir = "%s/ports/%u/" % (nsimdev.dfs_dir, port_index) ret = ip("-j link show dev %s" % ifname, ns=ns) self.dev = json.loads(ret.stdout)[0] + self.ifindex = self.dev["ifindex"] def dfs_write(self, path, val): self.nsimdev.dfs_write(f'ports/{self.port_index}/' + path, val) diff --git a/tools/testing/selftests/net/nl_netdev.py b/tools/testing/selftests/net/nl_netdev.py index 2b8b488fb419..6909b1760739 100755 --- a/tools/testing/selftests/net/nl_netdev.py +++ b/tools/testing/selftests/net/nl_netdev.py @@ -1,7 +1,9 @@ #!/usr/bin/env python3 # SPDX-License-Identifier: GPL-2.0 -from lib.py import ksft_run, ksft_pr, ksft_eq, ksft_ge, NetdevFamily +import time +from lib.py import ksft_run, ksft_pr, ksft_eq, ksft_ge, ksft_busy_wait +from lib.py import NetdevFamily, NetdevSimDev, ip def empty_check(nf) -> None: @@ -15,9 +17,79 @@ def lo_check(nf) -> None: ksft_eq(len(lo_info['xdp-rx-metadata-features']), 0) +def page_pool_check(nf) -> None: + with NetdevSimDev() as nsimdev: + nsim = nsimdev.nsims[0] + + def up(): + ip(f"link set dev {nsim.ifname} up") + + def down(): + ip(f"link set dev {nsim.ifname} down") + + def get_pp(): + pp_list = nf.page_pool_get({}, dump=True) + return [pp for pp in pp_list if pp.get("ifindex") == nsim.ifindex] + + # No page pools when down + down() + ksft_eq(len(get_pp()), 0) + + # Up, empty page pool appears + up() + pp_list = get_pp() + ksft_ge(len(pp_list), 0) + refs = sum([pp["inflight"] for pp in pp_list]) + ksft_eq(refs, 0) + + # Down, it disappears, again + down() + pp_list = get_pp() + ksft_eq(len(pp_list), 0) + + # Up, allocate a page + up() + nsim.dfs_write("pp_hold", "y") + pp_list = nf.page_pool_get({}, dump=True) + refs = sum([pp["inflight"] for pp in pp_list if pp.get("ifindex") == nsim.ifindex]) + ksft_ge(refs, 1) + + # Now let's leak a page + down() + pp_list = get_pp() + ksft_eq(len(pp_list), 1) + refs = sum([pp["inflight"] for pp in pp_list]) + ksft_eq(refs, 1) + attached = [pp for pp in pp_list if "detach-time" not in pp] + ksft_eq(len(attached), 0) + + # New pp can get created, and we'll have two + up() + pp_list = get_pp() + attached = [pp for pp in pp_list if "detach-time" not in pp] + detached = [pp for pp in pp_list if "detach-time" in pp] + ksft_eq(len(attached), 1) + ksft_eq(len(detached), 1) + + # Free the old page and the old pp is gone + nsim.dfs_write("pp_hold", "n") + # Freeing check is once a second so we may need to retry + ksft_busy_wait(lambda: len(get_pp()) == 1, deadline=2) + + # And down... + down() + ksft_eq(len(get_pp()), 0) + + # Last, leave the page hanging for destroy, nothing to check + # we're trying to exercise the orphaning path in the kernel + up() + nsim.dfs_write("pp_hold", "y") + + def main() -> None: nf = NetdevFamily() - ksft_run([empty_check, lo_check], args=(nf, )) + ksft_run([empty_check, lo_check, page_pool_check], + args=(nf, )) if __name__ == "__main__": From 0be9a1e43a072973736db1e6deb45836fcb9c02e Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 12 Apr 2024 16:15:13 +0100 Subject: [PATCH 11/25] net: dsa: sja1105: provide own phylink MAC operations Convert sja1105 to provide its own phylink MAC operations, thus avoiding the shim layer in DSA's port.c Signed-off-by: Russell King (Oracle) Reviewed-by: Florian Fainelli Link: https://lore.kernel.org/r/E1rvIcT-006bQW-S3@rmk-PC.armlinux.org.uk Signed-off-by: Paolo Abeni --- drivers/net/dsa/sja1105/sja1105_main.c | 38 ++++++++++++++++++-------- 1 file changed, 27 insertions(+), 11 deletions(-) diff --git a/drivers/net/dsa/sja1105/sja1105_main.c b/drivers/net/dsa/sja1105/sja1105_main.c index fc262348a134..ee0fb1c343f1 100644 --- a/drivers/net/dsa/sja1105/sja1105_main.c +++ b/drivers/net/dsa/sja1105/sja1105_main.c @@ -1358,10 +1358,11 @@ static int sja1105_adjust_port_config(struct sja1105_private *priv, int port, } static struct phylink_pcs * -sja1105_mac_select_pcs(struct dsa_switch *ds, int port, phy_interface_t iface) +sja1105_mac_select_pcs(struct phylink_config *config, phy_interface_t iface) { - struct sja1105_private *priv = ds->priv; - struct dw_xpcs *xpcs = priv->xpcs[port]; + struct dsa_port *dp = dsa_phylink_to_port(config); + struct sja1105_private *priv = dp->ds->priv; + struct dw_xpcs *xpcs = priv->xpcs[dp->index]; if (xpcs) return &xpcs->pcs; @@ -1369,21 +1370,31 @@ sja1105_mac_select_pcs(struct dsa_switch *ds, int port, phy_interface_t iface) return NULL; } -static void sja1105_mac_link_down(struct dsa_switch *ds, int port, +static void sja1105_mac_config(struct phylink_config *config, + unsigned int mode, + const struct phylink_link_state *state) +{ +} + +static void sja1105_mac_link_down(struct phylink_config *config, unsigned int mode, phy_interface_t interface) { - sja1105_inhibit_tx(ds->priv, BIT(port), true); + struct dsa_port *dp = dsa_phylink_to_port(config); + + sja1105_inhibit_tx(dp->ds->priv, BIT(dp->index), true); } -static void sja1105_mac_link_up(struct dsa_switch *ds, int port, +static void sja1105_mac_link_up(struct phylink_config *config, + struct phy_device *phydev, unsigned int mode, phy_interface_t interface, - struct phy_device *phydev, int speed, int duplex, bool tx_pause, bool rx_pause) { - struct sja1105_private *priv = ds->priv; + struct dsa_port *dp = dsa_phylink_to_port(config); + struct sja1105_private *priv = dp->ds->priv; + int port = dp->index; sja1105_adjust_port_config(priv, port, speed); @@ -3198,6 +3209,13 @@ static void sja1105_teardown(struct dsa_switch *ds) sja1105_static_config_free(&priv->static_config); } +static const struct phylink_mac_ops sja1105_phylink_mac_ops = { + .mac_select_pcs = sja1105_mac_select_pcs, + .mac_config = sja1105_mac_config, + .mac_link_up = sja1105_mac_link_up, + .mac_link_down = sja1105_mac_link_down, +}; + static const struct dsa_switch_ops sja1105_switch_ops = { .get_tag_protocol = sja1105_get_tag_protocol, .connect_tag_protocol = sja1105_connect_tag_protocol, @@ -3207,9 +3225,6 @@ static const struct dsa_switch_ops sja1105_switch_ops = { .port_change_mtu = sja1105_change_mtu, .port_max_mtu = sja1105_get_max_mtu, .phylink_get_caps = sja1105_phylink_get_caps, - .phylink_mac_select_pcs = sja1105_mac_select_pcs, - .phylink_mac_link_up = sja1105_mac_link_up, - .phylink_mac_link_down = sja1105_mac_link_down, .get_strings = sja1105_get_strings, .get_ethtool_stats = sja1105_get_ethtool_stats, .get_sset_count = sja1105_get_sset_count, @@ -3375,6 +3390,7 @@ static int sja1105_probe(struct spi_device *spi) ds->dev = dev; ds->num_ports = priv->info->num_ports; ds->ops = &sja1105_switch_ops; + ds->phylink_mac_ops = &sja1105_phylink_mac_ops; ds->priv = priv; priv->ds = ds; From e3ef87ef403e84101f365dc13e2e2d75c9c94415 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 12 Apr 2024 16:15:19 +0100 Subject: [PATCH 12/25] net: dsa: ar9331: provide own phylink MAC operations Convert ar9331 to provide its own phylink MAC operations, thus avoiding the shim layer in DSA's port.c. Signed-off-by: Russell King (Oracle) Reviewed-by: Vladimir Oltean Reviewed-by: Florian Fainelli Link: https://lore.kernel.org/r/E1rvIcZ-006bQc-0W@rmk-PC.armlinux.org.uk Signed-off-by: Paolo Abeni --- drivers/net/dsa/qca/ar9331.c | 37 +++++++++++++++++++++--------------- 1 file changed, 22 insertions(+), 15 deletions(-) diff --git a/drivers/net/dsa/qca/ar9331.c b/drivers/net/dsa/qca/ar9331.c index 8d9d271ac3af..968cb81088bf 100644 --- a/drivers/net/dsa/qca/ar9331.c +++ b/drivers/net/dsa/qca/ar9331.c @@ -523,28 +523,30 @@ static void ar9331_sw_phylink_get_caps(struct dsa_switch *ds, int port, } } -static void ar9331_sw_phylink_mac_config(struct dsa_switch *ds, int port, +static void ar9331_sw_phylink_mac_config(struct phylink_config *config, unsigned int mode, const struct phylink_link_state *state) { - struct ar9331_sw_priv *priv = ds->priv; + struct dsa_port *dp = dsa_phylink_to_port(config); + struct ar9331_sw_priv *priv = dp->ds->priv; struct regmap *regmap = priv->regmap; int ret; - ret = regmap_update_bits(regmap, AR9331_SW_REG_PORT_STATUS(port), + ret = regmap_update_bits(regmap, AR9331_SW_REG_PORT_STATUS(dp->index), AR9331_SW_PORT_STATUS_LINK_EN | AR9331_SW_PORT_STATUS_FLOW_LINK_EN, 0); if (ret) dev_err_ratelimited(priv->dev, "%s: %i\n", __func__, ret); } -static void ar9331_sw_phylink_mac_link_down(struct dsa_switch *ds, int port, +static void ar9331_sw_phylink_mac_link_down(struct phylink_config *config, unsigned int mode, phy_interface_t interface) { - struct ar9331_sw_priv *priv = ds->priv; - struct ar9331_sw_port *p = &priv->port[port]; + struct dsa_port *dp = dsa_phylink_to_port(config); + struct ar9331_sw_priv *priv = dp->ds->priv; struct regmap *regmap = priv->regmap; + int port = dp->index; int ret; ret = regmap_update_bits(regmap, AR9331_SW_REG_PORT_STATUS(port), @@ -552,23 +554,24 @@ static void ar9331_sw_phylink_mac_link_down(struct dsa_switch *ds, int port, if (ret) dev_err_ratelimited(priv->dev, "%s: %i\n", __func__, ret); - cancel_delayed_work_sync(&p->mib_read); + cancel_delayed_work_sync(&priv->port[port].mib_read); } -static void ar9331_sw_phylink_mac_link_up(struct dsa_switch *ds, int port, +static void ar9331_sw_phylink_mac_link_up(struct phylink_config *config, + struct phy_device *phydev, unsigned int mode, phy_interface_t interface, - struct phy_device *phydev, int speed, int duplex, bool tx_pause, bool rx_pause) { - struct ar9331_sw_priv *priv = ds->priv; - struct ar9331_sw_port *p = &priv->port[port]; + struct dsa_port *dp = dsa_phylink_to_port(config); + struct ar9331_sw_priv *priv = dp->ds->priv; struct regmap *regmap = priv->regmap; + int port = dp->index; u32 val; int ret; - schedule_delayed_work(&p->mib_read, 0); + schedule_delayed_work(&priv->port[port].mib_read, 0); val = AR9331_SW_PORT_STATUS_MAC_MASK; switch (speed) { @@ -684,14 +687,17 @@ static void ar9331_get_pause_stats(struct dsa_switch *ds, int port, spin_unlock(&p->stats_lock); } +static const struct phylink_mac_ops ar9331_phylink_mac_ops = { + .mac_config = ar9331_sw_phylink_mac_config, + .mac_link_down = ar9331_sw_phylink_mac_link_down, + .mac_link_up = ar9331_sw_phylink_mac_link_up, +}; + static const struct dsa_switch_ops ar9331_sw_ops = { .get_tag_protocol = ar9331_sw_get_tag_protocol, .setup = ar9331_sw_setup, .port_disable = ar9331_sw_port_disable, .phylink_get_caps = ar9331_sw_phylink_get_caps, - .phylink_mac_config = ar9331_sw_phylink_mac_config, - .phylink_mac_link_down = ar9331_sw_phylink_mac_link_down, - .phylink_mac_link_up = ar9331_sw_phylink_mac_link_up, .get_stats64 = ar9331_get_stats64, .get_pause_stats = ar9331_get_pause_stats, }; @@ -1059,6 +1065,7 @@ static int ar9331_sw_probe(struct mdio_device *mdiodev) ds->priv = priv; priv->ops = ar9331_sw_ops; ds->ops = &priv->ops; + ds->phylink_mac_ops = &ar9331_phylink_mac_ops; dev_set_drvdata(&mdiodev->dev, priv); for (i = 0; i < ARRAY_SIZE(priv->port); i++) { From 636d022cd586b94aa4334e1d9b2558f821d58c2e Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 12 Apr 2024 16:15:24 +0100 Subject: [PATCH 13/25] net: dsa: qca8k: provide own phylink MAC operations Convert qca8k to provide its own phylink MAC operations, thus avoiding the shim layer in DSA's port.c. Signed-off-by: Russell King (Oracle) Reviewed-by: Vladimir Oltean Reviewed-by: Florian Fainelli Link: https://lore.kernel.org/r/E1rvIce-006bQi-58@rmk-PC.armlinux.org.uk Signed-off-by: Paolo Abeni --- drivers/net/dsa/qca/qca8k-8xxx.c | 49 +++++++++++++++++++++----------- 1 file changed, 32 insertions(+), 17 deletions(-) diff --git a/drivers/net/dsa/qca/qca8k-8xxx.c b/drivers/net/dsa/qca/qca8k-8xxx.c index dab66c0c6f64..b3c27cf538e8 100644 --- a/drivers/net/dsa/qca/qca8k-8xxx.c +++ b/drivers/net/dsa/qca/qca8k-8xxx.c @@ -1283,11 +1283,13 @@ qca8k_mac_config_setup_internal_delay(struct qca8k_priv *priv, int cpu_port_inde } static struct phylink_pcs * -qca8k_phylink_mac_select_pcs(struct dsa_switch *ds, int port, +qca8k_phylink_mac_select_pcs(struct phylink_config *config, phy_interface_t interface) { - struct qca8k_priv *priv = ds->priv; + struct dsa_port *dp = dsa_phylink_to_port(config); + struct qca8k_priv *priv = dp->ds->priv; struct phylink_pcs *pcs = NULL; + int port = dp->index; switch (interface) { case PHY_INTERFACE_MODE_SGMII: @@ -1311,13 +1313,18 @@ qca8k_phylink_mac_select_pcs(struct dsa_switch *ds, int port, } static void -qca8k_phylink_mac_config(struct dsa_switch *ds, int port, unsigned int mode, +qca8k_phylink_mac_config(struct phylink_config *config, unsigned int mode, const struct phylink_link_state *state) { - struct qca8k_priv *priv = ds->priv; + struct dsa_port *dp = dsa_phylink_to_port(config); + struct dsa_switch *ds = dp->ds; + struct qca8k_priv *priv; + int port = dp->index; int cpu_port_index; u32 reg; + priv = ds->priv; + switch (port) { case 0: /* 1st CPU port */ if (state->interface != PHY_INTERFACE_MODE_RGMII && @@ -1426,20 +1433,24 @@ static void qca8k_phylink_get_caps(struct dsa_switch *ds, int port, } static void -qca8k_phylink_mac_link_down(struct dsa_switch *ds, int port, unsigned int mode, +qca8k_phylink_mac_link_down(struct phylink_config *config, unsigned int mode, phy_interface_t interface) { - struct qca8k_priv *priv = ds->priv; + struct dsa_port *dp = dsa_phylink_to_port(config); + struct qca8k_priv *priv = dp->ds->priv; - qca8k_port_set_status(priv, port, 0); + qca8k_port_set_status(priv, dp->index, 0); } static void -qca8k_phylink_mac_link_up(struct dsa_switch *ds, int port, unsigned int mode, - phy_interface_t interface, struct phy_device *phydev, - int speed, int duplex, bool tx_pause, bool rx_pause) +qca8k_phylink_mac_link_up(struct phylink_config *config, + struct phy_device *phydev, unsigned int mode, + phy_interface_t interface, int speed, int duplex, + bool tx_pause, bool rx_pause) { - struct qca8k_priv *priv = ds->priv; + struct dsa_port *dp = dsa_phylink_to_port(config); + struct qca8k_priv *priv = dp->ds->priv; + int port = dp->index; u32 reg; if (phylink_autoneg_inband(mode)) { @@ -1463,10 +1474,10 @@ qca8k_phylink_mac_link_up(struct dsa_switch *ds, int port, unsigned int mode, if (duplex == DUPLEX_FULL) reg |= QCA8K_PORT_STATUS_DUPLEX; - if (rx_pause || dsa_is_cpu_port(ds, port)) + if (rx_pause || dsa_port_is_cpu(dp)) reg |= QCA8K_PORT_STATUS_RXFLOW; - if (tx_pause || dsa_is_cpu_port(ds, port)) + if (tx_pause || dsa_port_is_cpu(dp)) reg |= QCA8K_PORT_STATUS_TXFLOW; } @@ -1991,6 +2002,13 @@ qca8k_setup(struct dsa_switch *ds) return 0; } +static const struct phylink_mac_ops qca8k_phylink_mac_ops = { + .mac_select_pcs = qca8k_phylink_mac_select_pcs, + .mac_config = qca8k_phylink_mac_config, + .mac_link_down = qca8k_phylink_mac_link_down, + .mac_link_up = qca8k_phylink_mac_link_up, +}; + static const struct dsa_switch_ops qca8k_switch_ops = { .get_tag_protocol = qca8k_get_tag_protocol, .setup = qca8k_setup, @@ -2021,10 +2039,6 @@ static const struct dsa_switch_ops qca8k_switch_ops = { .port_vlan_add = qca8k_port_vlan_add, .port_vlan_del = qca8k_port_vlan_del, .phylink_get_caps = qca8k_phylink_get_caps, - .phylink_mac_select_pcs = qca8k_phylink_mac_select_pcs, - .phylink_mac_config = qca8k_phylink_mac_config, - .phylink_mac_link_down = qca8k_phylink_mac_link_down, - .phylink_mac_link_up = qca8k_phylink_mac_link_up, .get_phy_flags = qca8k_get_phy_flags, .port_lag_join = qca8k_port_lag_join, .port_lag_leave = qca8k_port_lag_leave, @@ -2091,6 +2105,7 @@ qca8k_sw_probe(struct mdio_device *mdiodev) priv->ds->num_ports = QCA8K_NUM_PORTS; priv->ds->priv = priv; priv->ds->ops = &qca8k_switch_ops; + priv->ds->phylink_mac_ops = &qca8k_phylink_mac_ops; mutex_init(&priv->reg_mutex); dev_set_drvdata(&mdiodev->dev, priv); From 94c437edce65ff8cf71bf8b22801c81f48eefdd6 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 12 Apr 2024 16:15:29 +0100 Subject: [PATCH 14/25] net: dsa: lantiq_gswip: provide own phylink MAC operations Convert lantiq_gswip to provide its own phylink MAC operations, thus avoiding the shim layer in DSA's port.c. For lantiq_gswip, it means we end up with a common instance of phylink MAC operations that are shared between the different variants, rather than having duplicated initialisers in dsa_switch_ops. Signed-off-by: Russell King (Oracle) Link: https://lore.kernel.org/r/E1rvIcj-006bQo-B3@rmk-PC.armlinux.org.uk Signed-off-by: Paolo Abeni --- drivers/net/dsa/lantiq_gswip.c | 39 ++++++++++++++++++++-------------- 1 file changed, 23 insertions(+), 16 deletions(-) diff --git a/drivers/net/dsa/lantiq_gswip.c b/drivers/net/dsa/lantiq_gswip.c index de48b194048f..a557049e34f5 100644 --- a/drivers/net/dsa/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq_gswip.c @@ -1670,11 +1670,13 @@ static void gswip_port_set_pause(struct gswip_priv *priv, int port, mdio_phy, GSWIP_MDIO_PHYp(port)); } -static void gswip_phylink_mac_config(struct dsa_switch *ds, int port, +static void gswip_phylink_mac_config(struct phylink_config *config, unsigned int mode, const struct phylink_link_state *state) { - struct gswip_priv *priv = ds->priv; + struct dsa_port *dp = dsa_phylink_to_port(config); + struct gswip_priv *priv = dp->ds->priv; + int port = dp->index; u32 miicfg = 0; miicfg |= GSWIP_MII_CFG_LDCLKDIS; @@ -1700,7 +1702,7 @@ static void gswip_phylink_mac_config(struct dsa_switch *ds, int port, miicfg |= GSWIP_MII_CFG_MODE_GMII; break; default: - dev_err(ds->dev, + dev_err(dp->ds->dev, "Unsupported interface: %d\n", state->interface); return; } @@ -1726,28 +1728,32 @@ static void gswip_phylink_mac_config(struct dsa_switch *ds, int port, } } -static void gswip_phylink_mac_link_down(struct dsa_switch *ds, int port, +static void gswip_phylink_mac_link_down(struct phylink_config *config, unsigned int mode, phy_interface_t interface) { - struct gswip_priv *priv = ds->priv; + struct dsa_port *dp = dsa_phylink_to_port(config); + struct gswip_priv *priv = dp->ds->priv; + int port = dp->index; gswip_mii_mask_cfg(priv, GSWIP_MII_CFG_EN, 0, port); - if (!dsa_is_cpu_port(ds, port)) + if (!dsa_port_is_cpu(dp)) gswip_port_set_link(priv, port, false); } -static void gswip_phylink_mac_link_up(struct dsa_switch *ds, int port, +static void gswip_phylink_mac_link_up(struct phylink_config *config, + struct phy_device *phydev, unsigned int mode, phy_interface_t interface, - struct phy_device *phydev, int speed, int duplex, bool tx_pause, bool rx_pause) { - struct gswip_priv *priv = ds->priv; + struct dsa_port *dp = dsa_phylink_to_port(config); + struct gswip_priv *priv = dp->ds->priv; + int port = dp->index; - if (!dsa_is_cpu_port(ds, port)) { + if (!dsa_port_is_cpu(dp)) { gswip_port_set_link(priv, port, true); gswip_port_set_speed(priv, port, speed, interface); gswip_port_set_duplex(priv, port, duplex); @@ -1824,6 +1830,12 @@ static int gswip_get_sset_count(struct dsa_switch *ds, int port, int sset) return ARRAY_SIZE(gswip_rmon_cnt); } +static const struct phylink_mac_ops gswip_phylink_mac_ops = { + .mac_config = gswip_phylink_mac_config, + .mac_link_down = gswip_phylink_mac_link_down, + .mac_link_up = gswip_phylink_mac_link_up, +}; + static const struct dsa_switch_ops gswip_xrx200_switch_ops = { .get_tag_protocol = gswip_get_tag_protocol, .setup = gswip_setup, @@ -1842,9 +1854,6 @@ static const struct dsa_switch_ops gswip_xrx200_switch_ops = { .port_change_mtu = gswip_port_change_mtu, .port_max_mtu = gswip_port_max_mtu, .phylink_get_caps = gswip_xrx200_phylink_get_caps, - .phylink_mac_config = gswip_phylink_mac_config, - .phylink_mac_link_down = gswip_phylink_mac_link_down, - .phylink_mac_link_up = gswip_phylink_mac_link_up, .get_strings = gswip_get_strings, .get_ethtool_stats = gswip_get_ethtool_stats, .get_sset_count = gswip_get_sset_count, @@ -1868,9 +1877,6 @@ static const struct dsa_switch_ops gswip_xrx300_switch_ops = { .port_change_mtu = gswip_port_change_mtu, .port_max_mtu = gswip_port_max_mtu, .phylink_get_caps = gswip_xrx300_phylink_get_caps, - .phylink_mac_config = gswip_phylink_mac_config, - .phylink_mac_link_down = gswip_phylink_mac_link_down, - .phylink_mac_link_up = gswip_phylink_mac_link_up, .get_strings = gswip_get_strings, .get_ethtool_stats = gswip_get_ethtool_stats, .get_sset_count = gswip_get_sset_count, @@ -2136,6 +2142,7 @@ static int gswip_probe(struct platform_device *pdev) priv->ds->num_ports = priv->hw_info->max_ports; priv->ds->priv = priv; priv->ds->ops = priv->hw_info->ops; + priv->ds->phylink_mac_ops = &gswip_phylink_mac_ops; priv->dev = dev; mutex_init(&priv->pce_table_lock); version = gswip_switch_r(priv, GSWIP_VERSION); From 62d6d91db98a04defffd4ecfe168dd6ca80ef7d1 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 12 Apr 2024 16:15:34 +0100 Subject: [PATCH 15/25] net: dsa: mt7530: provide own phylink MAC operations MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Convert mt753x to provide its own phylink MAC operations, thus avoiding the shim layer in DSA's port.c Signed-off-by: Russell King (Oracle) Tested-by: Arınç ÜNAL Link: https://lore.kernel.org/r/E1rvIco-006bQu-Fq@rmk-PC.armlinux.org.uk Signed-off-by: Paolo Abeni --- drivers/net/dsa/mt7530.c | 46 +++++++++++++++++++++++++--------------- 1 file changed, 29 insertions(+), 17 deletions(-) diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c index c0d0bce0b594..84eec7de9b03 100644 --- a/drivers/net/dsa/mt7530.c +++ b/drivers/net/dsa/mt7530.c @@ -2850,28 +2850,34 @@ mt7531_mac_config(struct dsa_switch *ds, int port, unsigned int mode, } static struct phylink_pcs * -mt753x_phylink_mac_select_pcs(struct dsa_switch *ds, int port, +mt753x_phylink_mac_select_pcs(struct phylink_config *config, phy_interface_t interface) { - struct mt7530_priv *priv = ds->priv; + struct dsa_port *dp = dsa_phylink_to_port(config); + struct mt7530_priv *priv = dp->ds->priv; switch (interface) { case PHY_INTERFACE_MODE_TRGMII: - return &priv->pcs[port].pcs; + return &priv->pcs[dp->index].pcs; case PHY_INTERFACE_MODE_SGMII: case PHY_INTERFACE_MODE_1000BASEX: case PHY_INTERFACE_MODE_2500BASEX: - return priv->ports[port].sgmii_pcs; + return priv->ports[dp->index].sgmii_pcs; default: return NULL; } } static void -mt753x_phylink_mac_config(struct dsa_switch *ds, int port, unsigned int mode, +mt753x_phylink_mac_config(struct phylink_config *config, unsigned int mode, const struct phylink_link_state *state) { - struct mt7530_priv *priv = ds->priv; + struct dsa_port *dp = dsa_phylink_to_port(config); + struct dsa_switch *ds = dp->ds; + struct mt7530_priv *priv; + int port = dp->index; + + priv = ds->priv; if ((port == 5 || port == 6) && priv->info->mac_port_config) priv->info->mac_port_config(ds, port, mode, state->interface); @@ -2881,23 +2887,25 @@ mt753x_phylink_mac_config(struct dsa_switch *ds, int port, unsigned int mode, mt7530_set(priv, MT7530_PMCR_P(port), PMCR_EXT_PHY); } -static void mt753x_phylink_mac_link_down(struct dsa_switch *ds, int port, +static void mt753x_phylink_mac_link_down(struct phylink_config *config, unsigned int mode, phy_interface_t interface) { - struct mt7530_priv *priv = ds->priv; + struct dsa_port *dp = dsa_phylink_to_port(config); + struct mt7530_priv *priv = dp->ds->priv; - mt7530_clear(priv, MT7530_PMCR_P(port), PMCR_LINK_SETTINGS_MASK); + mt7530_clear(priv, MT7530_PMCR_P(dp->index), PMCR_LINK_SETTINGS_MASK); } -static void mt753x_phylink_mac_link_up(struct dsa_switch *ds, int port, +static void mt753x_phylink_mac_link_up(struct phylink_config *config, + struct phy_device *phydev, unsigned int mode, phy_interface_t interface, - struct phy_device *phydev, int speed, int duplex, bool tx_pause, bool rx_pause) { - struct mt7530_priv *priv = ds->priv; + struct dsa_port *dp = dsa_phylink_to_port(config); + struct mt7530_priv *priv = dp->ds->priv; u32 mcr; mcr = PMCR_RX_EN | PMCR_TX_EN | PMCR_FORCE_LNK; @@ -2932,7 +2940,7 @@ static void mt753x_phylink_mac_link_up(struct dsa_switch *ds, int port, } } - mt7530_set(priv, MT7530_PMCR_P(port), mcr); + mt7530_set(priv, MT7530_PMCR_P(dp->index), mcr); } static void mt753x_phylink_get_caps(struct dsa_switch *ds, int port, @@ -3152,16 +3160,19 @@ const struct dsa_switch_ops mt7530_switch_ops = { .port_mirror_add = mt753x_port_mirror_add, .port_mirror_del = mt753x_port_mirror_del, .phylink_get_caps = mt753x_phylink_get_caps, - .phylink_mac_select_pcs = mt753x_phylink_mac_select_pcs, - .phylink_mac_config = mt753x_phylink_mac_config, - .phylink_mac_link_down = mt753x_phylink_mac_link_down, - .phylink_mac_link_up = mt753x_phylink_mac_link_up, .get_mac_eee = mt753x_get_mac_eee, .set_mac_eee = mt753x_set_mac_eee, .conduit_state_change = mt753x_conduit_state_change, }; EXPORT_SYMBOL_GPL(mt7530_switch_ops); +static const struct phylink_mac_ops mt753x_phylink_mac_ops = { + .mac_select_pcs = mt753x_phylink_mac_select_pcs, + .mac_config = mt753x_phylink_mac_config, + .mac_link_down = mt753x_phylink_mac_link_down, + .mac_link_up = mt753x_phylink_mac_link_up, +}; + const struct mt753x_info mt753x_table[] = { [ID_MT7621] = { .id = ID_MT7621, @@ -3239,6 +3250,7 @@ mt7530_probe_common(struct mt7530_priv *priv) priv->dev = dev; priv->ds->priv = priv; priv->ds->ops = &mt7530_switch_ops; + priv->ds->phylink_mac_ops = &mt753x_phylink_mac_ops; mutex_init(&priv->reg_mutex); dev_set_drvdata(dev, priv); From a4022a332f437ae5b10921d66058ce98a2db2c20 Mon Sep 17 00:00:00 2001 From: Petr Machata Date: Fri, 12 Apr 2024 19:03:04 +0200 Subject: [PATCH 16/25] selftests: net: Unify code of busywait() and slowwait() Bodies of busywait() and slowwait() functions are almost identical. Extract the common code into a helper, loopy_wait, and convert busywait() and slowwait() into trivial wrappers. Moreover, the fact that slowwait() uses seconds for units is really not intuitive, and the comment does not help much. Instead make the unit part of the name of the argument to further clarify what units are expected. Cc: Hangbin Liu Signed-off-by: Petr Machata Reviewed-by: Benjamin Poirier Reviewed-by: Hangbin Liu Signed-off-by: Paolo Abeni --- tools/testing/selftests/net/forwarding/lib.sh | 22 ++----------------- tools/testing/selftests/net/lib.sh | 16 +++++++++++--- 2 files changed, 15 insertions(+), 23 deletions(-) diff --git a/tools/testing/selftests/net/forwarding/lib.sh b/tools/testing/selftests/net/forwarding/lib.sh index 4103ed7afcde..658e4e7bf4b9 100644 --- a/tools/testing/selftests/net/forwarding/lib.sh +++ b/tools/testing/selftests/net/forwarding/lib.sh @@ -95,27 +95,9 @@ source "$net_forwarding_dir/../lib.sh" # timeout in seconds slowwait() { - local timeout=$1; shift + local timeout_sec=$1; shift - local start_time="$(date -u +%s)" - while true - do - local out - out=$("$@") - local ret=$? - if ((!ret)); then - echo -n "$out" - return 0 - fi - - local current_time="$(date -u +%s)" - if ((current_time - start_time > timeout)); then - echo -n "$out" - return 1 - fi - - sleep 0.1 - done + loopy_wait "sleep 0.1" "$((timeout_sec * 1000))" "$@" } ############################################################################## diff --git a/tools/testing/selftests/net/lib.sh b/tools/testing/selftests/net/lib.sh index b7f7b8695165..c868c0aec121 100644 --- a/tools/testing/selftests/net/lib.sh +++ b/tools/testing/selftests/net/lib.sh @@ -58,9 +58,10 @@ ksft_exit_status_merge() $ksft_xfail $ksft_pass $ksft_skip $ksft_fail } -busywait() +loopy_wait() { - local timeout=$1; shift + local sleep_cmd=$1; shift + local timeout_ms=$1; shift local start_time="$(date -u +%s%3N)" while true @@ -74,13 +75,22 @@ busywait() fi local current_time="$(date -u +%s%3N)" - if ((current_time - start_time > timeout)); then + if ((current_time - start_time > timeout_ms)); then echo -n "$out" return 1 fi + + $sleep_cmd done } +busywait() +{ + local timeout_ms=$1; shift + + loopy_wait : "$timeout_ms" "$@" +} + cleanup_ns() { local ns="" From 2291752fae3d2d773f2d29c159d383138411d06f Mon Sep 17 00:00:00 2001 From: Petr Machata Date: Fri, 12 Apr 2024 19:03:05 +0200 Subject: [PATCH 17/25] selftests: forwarding: lib.sh: Validate NETIFS The variable should contain at least NUM_NETIFS interfaces, stored as keys named "p$i", for i in `seq $NUM_NETIFS`. Signed-off-by: Petr Machata Reviewed-by: Benjamin Poirier Reviewed-by: Hangbin Liu Signed-off-by: Paolo Abeni --- tools/testing/selftests/net/forwarding/lib.sh | 22 ++++++++++++++----- 1 file changed, 17 insertions(+), 5 deletions(-) diff --git a/tools/testing/selftests/net/forwarding/lib.sh b/tools/testing/selftests/net/forwarding/lib.sh index 658e4e7bf4b9..3cbbc2fd4d7d 100644 --- a/tools/testing/selftests/net/forwarding/lib.sh +++ b/tools/testing/selftests/net/forwarding/lib.sh @@ -273,11 +273,6 @@ if [[ "$REQUIRE_MTOOLS" = "yes" ]]; then require_command mreceive fi -if [[ ! -v NUM_NETIFS ]]; then - echo "SKIP: importer does not define \"NUM_NETIFS\"" - exit $ksft_skip -fi - ############################################################################## # Command line options handling @@ -296,6 +291,23 @@ done ############################################################################## # Network interfaces configuration +if [[ ! -v NUM_NETIFS ]]; then + echo "SKIP: importer does not define \"NUM_NETIFS\"" + exit $ksft_skip +fi + +if (( NUM_NETIFS > ${#NETIFS[@]} )); then + echo "SKIP: Importer requires $NUM_NETIFS NETIFS, but only ${#NETIFS[@]} are defined (${NETIFS[@]})" + exit $ksft_skip +fi + +for i in $(seq ${#NETIFS[@]}); do + if [[ ! ${NETIFS[p$i]} ]]; then + echo "SKIP: NETIFS[p$i] not given" + exit $ksft_skip + fi +done + create_netif_veth() { local i From 492976136bb9fac0ebda28a163561eea8c2896d5 Mon Sep 17 00:00:00 2001 From: Petr Machata Date: Fri, 12 Apr 2024 19:03:06 +0200 Subject: [PATCH 18/25] selftests: forwarding: bail_on_lldpad() should SKIP $ksft_skip is used to mark selftests that have tooling issues. The fact that LLDPad is running, but shouldn't, is one such issue. Therefore have bail_on_lldpad() bail with $ksft_skip. Signed-off-by: Petr Machata Reviewed-by: Benjamin Poirier Reviewed-by: Hangbin Liu Signed-off-by: Paolo Abeni --- tools/testing/selftests/net/forwarding/lib.sh | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/net/forwarding/lib.sh b/tools/testing/selftests/net/forwarding/lib.sh index 3cbbc2fd4d7d..7913c6ee418d 100644 --- a/tools/testing/selftests/net/forwarding/lib.sh +++ b/tools/testing/selftests/net/forwarding/lib.sh @@ -2138,6 +2138,8 @@ bail_on_lldpad() { local reason1="$1"; shift local reason2="$1"; shift + local caller=${FUNCNAME[1]} + local src=${BASH_SOURCE[1]} if systemctl is-active --quiet lldpad; then @@ -2158,7 +2160,8 @@ bail_on_lldpad() an environment variable ALLOW_LLDPAD to a non-empty string. EOF - exit 1 + log_test_skip $src:$caller + exit $EXIT_STATUS else return fi From 042db639bf33aee118846a87223acebe6700137d Mon Sep 17 00:00:00 2001 From: Petr Machata Date: Fri, 12 Apr 2024 19:03:07 +0200 Subject: [PATCH 19/25] selftests: drivers: hw: Fix ethtool_rmon When rx-pktsNtoM reports a range that involves very low-valued range, such as 0-64, the calculated length of the packet will be -4, because FCS is subtracted from the value. mausezahn then confuses the value for an option and bails out. As a result, the test dumps many mausezahn error messages. Instead, cap the value at 0. mausezahn will use an appropriate minimum packet length. Cc: Vladimir Oltean Cc: Tobias Waldekranz Signed-off-by: Petr Machata Signed-off-by: Paolo Abeni --- tools/testing/selftests/drivers/net/hw/ethtool_rmon.sh | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/testing/selftests/drivers/net/hw/ethtool_rmon.sh b/tools/testing/selftests/drivers/net/hw/ethtool_rmon.sh index e2a1c10d3503..8f60c1685ad4 100755 --- a/tools/testing/selftests/drivers/net/hw/ethtool_rmon.sh +++ b/tools/testing/selftests/drivers/net/hw/ethtool_rmon.sh @@ -44,6 +44,7 @@ bucket_test() # Mausezahn does not include FCS bytes in its length - but the # histogram counters do len=$((len - ETH_FCS_LEN)) + len=$((len > 0 ? len : 0)) before=$(ethtool --json -S $iface --groups rmon | \ jq -r ".[0].rmon[\"${set}-pktsNtoM\"][$bucket].val") From f359d44a4e83eba434f00c4363e4b97928fd5661 Mon Sep 17 00:00:00 2001 From: Petr Machata Date: Fri, 12 Apr 2024 19:03:08 +0200 Subject: [PATCH 20/25] selftests: drivers: hw: ethtool.sh: Adjust output Some log_test calls are done in a loop, and lead to the same log output. This might prove tricky to deduplicate for automated tools. Instead, roll the unique information from log_info to log_test, and drop the log_info. This also leads to more compact and clearer output. This change prompts rewording the messages so that they are not excessively long. Some check_err messages do not indicate what the issue actually is, so reword them to say it's a "ping with", like is the case in some other instances in this test. Signed-off-by: Petr Machata Signed-off-by: Paolo Abeni --- tools/testing/selftests/drivers/net/hw/ethtool.sh | 15 ++++++--------- 1 file changed, 6 insertions(+), 9 deletions(-) diff --git a/tools/testing/selftests/drivers/net/hw/ethtool.sh b/tools/testing/selftests/drivers/net/hw/ethtool.sh index bb12d5d70949..fa6953de6b6d 100755 --- a/tools/testing/selftests/drivers/net/hw/ethtool.sh +++ b/tools/testing/selftests/drivers/net/hw/ethtool.sh @@ -65,9 +65,8 @@ same_speeds_autoneg_off() setup_wait_dev_with_timeout $h1 setup_wait_dev_with_timeout $h2 ping_do $h1 192.0.2.2 - check_err $? "speed $speed autoneg off" - log_test "force of same speed autoneg off" - log_info "speed = $speed" + check_err $? "ping with speed $speed autoneg off" + log_test "force speed $speed on both ends" done ethtool -s $h2 autoneg on @@ -112,9 +111,8 @@ combination_of_neg_on_and_off() setup_wait_dev_with_timeout $h1 setup_wait_dev_with_timeout $h2 ping_do $h1 192.0.2.2 - check_err $? "h1-speed=$speed autoneg off, h2 autoneg on" - log_test "one side with autoneg off and another with autoneg on" - log_info "force speed = $speed" + check_err $? "ping with h1-speed=$speed autoneg off, h2 autoneg on" + log_test "force speed $speed vs. autoneg" done ethtool -s $h1 autoneg on @@ -207,10 +205,9 @@ advertise_subset_of_speeds() setup_wait_dev_with_timeout $h1 setup_wait_dev_with_timeout $h2 ping_do $h1 192.0.2.2 - check_err $? "h1=$speed_1_to_advertise, h2=$speed_2_to_advertise ($speed_value)" + check_err $? "ping with h1=$speed_1_to_advertise, h2=$speed_2_to_advertise ($speed_value)" - log_test "advertise subset of speeds" - log_info "h1=$speed_1_to_advertise, h2=$speed_2_to_advertise" + log_test "advertise $speed_1_to_advertise vs. $speed_2_to_advertise" done ethtool -s $h2 autoneg on From bfc42940682b809e1f51491d8d9f82b7638fcbde Mon Sep 17 00:00:00 2001 From: Petr Machata Date: Fri, 12 Apr 2024 19:03:09 +0200 Subject: [PATCH 21/25] selftests: drivers: hw: Include tc_common.sh in hw_stats_l3 The tests use the constant TC_HIT_TIMEOUT when waiting on the counter values. However it does not include tc_common.sh where the counter is specified. The test has been robust in our testing, which means the counter is bumped quickly enough that the updated value is available already on the first iteration. Nevertheless it's not correct. Include tc_common.sh as appropriate. Signed-off-by: Petr Machata Signed-off-by: Paolo Abeni --- tools/testing/selftests/drivers/net/hw/hw_stats_l3.sh | 1 + tools/testing/selftests/drivers/net/hw/hw_stats_l3_gre.sh | 1 + 2 files changed, 2 insertions(+) diff --git a/tools/testing/selftests/drivers/net/hw/hw_stats_l3.sh b/tools/testing/selftests/drivers/net/hw/hw_stats_l3.sh index 7dfc50366c99..67fafefc80be 100755 --- a/tools/testing/selftests/drivers/net/hw/hw_stats_l3.sh +++ b/tools/testing/selftests/drivers/net/hw/hw_stats_l3.sh @@ -50,6 +50,7 @@ ALL_TESTS=" NUM_NETIFS=4 lib_dir=$(dirname "$0") source "$lib_dir"/../../../net/forwarding/lib.sh +source "$lib_dir"/../../../net/forwarding/tc_common.sh h1_create() { diff --git a/tools/testing/selftests/drivers/net/hw/hw_stats_l3_gre.sh b/tools/testing/selftests/drivers/net/hw/hw_stats_l3_gre.sh index ab8d04855af5..a94d92e1abce 100755 --- a/tools/testing/selftests/drivers/net/hw/hw_stats_l3_gre.sh +++ b/tools/testing/selftests/drivers/net/hw/hw_stats_l3_gre.sh @@ -15,6 +15,7 @@ NUM_NETIFS=6 lib_dir=$(dirname "$0") source "$lib_dir"/../../../net/forwarding/lib.sh source "$lib_dir"/../../../net/forwarding/ipip_lib.sh +source "$lib_dir"/../../../net/forwarding/tc_common.sh setup_prepare() { From 8d612ed4b55435812fe1cfb73be1dff4a4b9ce07 Mon Sep 17 00:00:00 2001 From: Danielle Ratson Date: Fri, 12 Apr 2024 19:03:10 +0200 Subject: [PATCH 22/25] selftests: mlxsw: ethtool_lanes: Wait for lanes parameter dump explicitly The ethtool dump includes the lanes parameter only when the port is up. Therefore, the ethtool_lanes.sh test waits for ports to come before testing the lanes parameter. In some cases, the test considers the port as up, but the lanes parameter is not yet dumped although assumed to be, resulting in ethtool_lanes.sh test failure. To avoid that, ensure that the lanes parameter is indeed dumped by waiting for it explicitly, before preforming the test cases. Signed-off-by: Danielle Ratson Reviewed-by: Petr Machata Signed-off-by: Petr Machata Signed-off-by: Paolo Abeni --- .../selftests/drivers/net/mlxsw/ethtool_lanes.sh | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/tools/testing/selftests/drivers/net/mlxsw/ethtool_lanes.sh b/tools/testing/selftests/drivers/net/mlxsw/ethtool_lanes.sh index 91891b9418d7..877cd6df94a1 100755 --- a/tools/testing/selftests/drivers/net/mlxsw/ethtool_lanes.sh +++ b/tools/testing/selftests/drivers/net/mlxsw/ethtool_lanes.sh @@ -24,8 +24,8 @@ setup_prepare() busywait "$TIMEOUT" wait_for_port_up ethtool $swp2 check_err $? "ports did not come up" - local lanes_exist=$(ethtool $swp1 | grep 'Lanes:') - if [[ -z $lanes_exist ]]; then + busywait $TIMEOUT sh -c "ethtool $swp1 | grep -q Lanes:" + if [[ $? -ne 0 ]]; then log_test "SKIP: driver does not support lanes setting" exit 1 fi @@ -122,8 +122,9 @@ autoneg() ethtool_set $swp1 speed $max_speed lanes $lanes ip link set dev $swp1 up ip link set dev $swp2 up - busywait "$TIMEOUT" wait_for_port_up ethtool $swp2 - check_err $? "ports did not come up" + + busywait $TIMEOUT sh -c "ethtool $swp1 | grep -q Lanes:" + check_err $? "Lanes parameter is not presented on time" check_lanes $swp1 $lanes $max_speed log_test "$lanes lanes is autonegotiated" @@ -160,8 +161,9 @@ autoneg_force_mode() ethtool_set $swp2 speed $max_speed lanes $lanes autoneg off ip link set dev $swp1 up ip link set dev $swp2 up - busywait "$TIMEOUT" wait_for_port_up ethtool $swp2 - check_err $? "ports did not come up" + + busywait $TIMEOUT sh -c "ethtool $swp1 | grep -q Lanes:" + check_err $? "Lanes parameter is not presented on time" check_lanes $swp1 $lanes $max_speed log_test "Autoneg off, $lanes lanes detected during force mode" From ba7d1e99b1935b06429b531aafe18d4e3a04cf9d Mon Sep 17 00:00:00 2001 From: Petr Machata Date: Fri, 12 Apr 2024 19:03:11 +0200 Subject: [PATCH 23/25] selftests: forwarding: router_mpath_nh: Add a diagram This test lacks a topology diagram, making the setup not obvious. Add one. Cc: David Ahern Signed-off-by: Petr Machata Reviewed-by: Hangbin Liu Reviewed-by: David Ahern Signed-off-by: Paolo Abeni --- .../net/forwarding/router_mpath_nh.sh | 35 +++++++++++++++++++ 1 file changed, 35 insertions(+) diff --git a/tools/testing/selftests/net/forwarding/router_mpath_nh.sh b/tools/testing/selftests/net/forwarding/router_mpath_nh.sh index 3f0f5dc95542..2ba44247c60a 100755 --- a/tools/testing/selftests/net/forwarding/router_mpath_nh.sh +++ b/tools/testing/selftests/net/forwarding/router_mpath_nh.sh @@ -1,6 +1,41 @@ #!/bin/bash # SPDX-License-Identifier: GPL-2.0 +# +-------------------------+ +# | H1 | +# | $h1 + | +# | 192.0.2.2/24 | | +# | 2001:db8:1::2/64 | | +# +-------------------|-----+ +# | +# +-------------------|----------------------+ +# | | R1 | +# | $rp11 + | +# | 192.0.2.1/24 | +# | 2001:db8:1::1/64 | +# | | +# | + $rp12 + $rp13 | +# | | 169.254.2.12/24 | 169.254.3.13/24 | +# | | fe80:2::12/64 | fe80:3::13/64 | +# +--|--------------------|------------------+ +# | | +# +--|--------------------|------------------+ +# | + $rp22 + $rp23 | +# | 169.254.2.22/24 169.254.3.23/24 | +# | fe80:2::22/64 fe80:3::23/64 | +# | | +# | $rp21 + | +# | 198.51.100.1/24 | | +# | 2001:db8:2::1/64 | R2 | +# +-------------------|----------------------+ +# | +# +-------------------|-----+ +# | | | +# | $h2 + | +# | 198.51.100.2/24 | +# | 2001:db8:2::2/64 H2 | +# +-------------------------+ + ALL_TESTS=" ping_ipv4 ping_ipv6 From b51a94b2d59d8d5f34a5eba1a6b3e733d5cb4355 Mon Sep 17 00:00:00 2001 From: Petr Machata Date: Fri, 12 Apr 2024 19:03:12 +0200 Subject: [PATCH 24/25] selftests: forwarding: router_mpath_nh_res: Add a diagram This test lacks a topology diagram, making the setup not obvious. Add one. Signed-off-by: Petr Machata Reviewed-by: Hangbin Liu Signed-off-by: Paolo Abeni --- .../net/forwarding/router_mpath_nh_res.sh | 35 +++++++++++++++++++ 1 file changed, 35 insertions(+) diff --git a/tools/testing/selftests/net/forwarding/router_mpath_nh_res.sh b/tools/testing/selftests/net/forwarding/router_mpath_nh_res.sh index 4b483d24ad00..cd9e346436fc 100755 --- a/tools/testing/selftests/net/forwarding/router_mpath_nh_res.sh +++ b/tools/testing/selftests/net/forwarding/router_mpath_nh_res.sh @@ -1,6 +1,41 @@ #!/bin/bash # SPDX-License-Identifier: GPL-2.0 +# +-------------------------+ +# | H1 | +# | $h1 + | +# | 192.0.2.2/24 | | +# | 2001:db8:1::2/64 | | +# +-------------------|-----+ +# | +# +-------------------|----------------------+ +# | | R1 | +# | $rp11 + | +# | 192.0.2.1/24 | +# | 2001:db8:1::1/64 | +# | | +# | + $rp12 + $rp13 | +# | | 169.254.2.12/24 | 169.254.3.13/24 | +# | | fe80:2::12/64 | fe80:3::13/64 | +# +--|--------------------|------------------+ +# | | +# +--|--------------------|------------------+ +# | + $rp22 + $rp23 | +# | 169.254.2.22/24 169.254.3.23/24 | +# | fe80:2::22/64 fe80:3::23/64 | +# | | +# | $rp21 + | +# | 198.51.100.1/24 | | +# | 2001:db8:2::1/64 | R2 | +# +-------------------|----------------------+ +# | +# +-------------------|-----+ +# | | | +# | $h2 + | +# | 198.51.100.2/24 | +# | 2001:db8:2::2/64 H2 | +# +-------------------------+ + ALL_TESTS=" ping_ipv4 ping_ipv6 From 74ddac073cfe5f47c1509b665ec9b30df1fcd71c Mon Sep 17 00:00:00 2001 From: Petr Machata Date: Fri, 12 Apr 2024 19:03:13 +0200 Subject: [PATCH 25/25] selftests: forwarding: router_nh: Add a diagram This test lacks a topology diagram, making the setup not obvious. Add one. Signed-off-by: Petr Machata Reviewed-by: Hangbin Liu Signed-off-by: Paolo Abeni --- .../testing/selftests/net/forwarding/router_nh.sh | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/tools/testing/selftests/net/forwarding/router_nh.sh b/tools/testing/selftests/net/forwarding/router_nh.sh index f3a53738bdcc..92904b01eae9 100755 --- a/tools/testing/selftests/net/forwarding/router_nh.sh +++ b/tools/testing/selftests/net/forwarding/router_nh.sh @@ -1,6 +1,20 @@ #!/bin/bash # SPDX-License-Identifier: GPL-2.0 +# +-------------------------+ +-------------------------+ +# | H1 | | H2 | +# | $h1 + | | $h2 + | +# | 192.0.2.2/24 | | | 198.51.100.2/24 | | +# | 2001:db8:1::2/64 | | | 2001:db8:2::2/64 | | +# +-------------------|-----+ +-------------------|-----+ +# | | +# +-------------------|----------------------------|-----+ +# | R1 | | | +# | $rp1 + $rp2 + | +# | 192.0.2.1/24 198.51.100.1/24 | +# | 2001:db8:1::1/64 2001:db8:2::1/64 | +# +------------------------------------------------------+ + ALL_TESTS=" ping_ipv4 ping_ipv6