diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2020-10-16 04:42:13 +0300 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2020-10-16 04:42:13 +0300 |
commit | 9ff9b0d392ea08090cd1780fb196f36dbb586529 (patch) | |
tree | 276a3a5c4525b84dee64eda30b423fc31bf94850 /drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c | |
parent | 840e5bb326bbcb16ce82dd2416d2769de4839aea (diff) | |
parent | 105faa8742437c28815b2a3eb8314ebc5fd9288c (diff) | |
download | linux-9ff9b0d392ea08090cd1780fb196f36dbb586529.tar.xz |
Merge tag 'net-next-5.10' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next
Pull networking updates from Jakub Kicinski:
- Add redirect_neigh() BPF packet redirect helper, allowing to limit
stack traversal in common container configs and improving TCP
back-pressure.
Daniel reports ~10Gbps => ~15Gbps single stream TCP performance gain.
- Expand netlink policy support and improve policy export to user
space. (Ge)netlink core performs request validation according to
declared policies. Expand the expressiveness of those policies
(min/max length and bitmasks). Allow dumping policies for particular
commands. This is used for feature discovery by user space (instead
of kernel version parsing or trial and error).
- Support IGMPv3/MLDv2 multicast listener discovery protocols in
bridge.
- Allow more than 255 IPv4 multicast interfaces.
- Add support for Type of Service (ToS) reflection in SYN/SYN-ACK
packets of TCPv6.
- In Multi-patch TCP (MPTCP) support concurrent transmission of data on
multiple subflows in a load balancing scenario. Enhance advertising
addresses via the RM_ADDR/ADD_ADDR options.
- Support SMC-Dv2 version of SMC, which enables multi-subnet
deployments.
- Allow more calls to same peer in RxRPC.
- Support two new Controller Area Network (CAN) protocols - CAN-FD and
ISO 15765-2:2016.
- Add xfrm/IPsec compat layer, solving the 32bit user space on 64bit
kernel problem.
- Add TC actions for implementing MPLS L2 VPNs.
- Improve nexthop code - e.g. handle various corner cases when nexthop
objects are removed from groups better, skip unnecessary
notifications and make it easier to offload nexthops into HW by
converting to a blocking notifier.
- Support adding and consuming TCP header options by BPF programs,
opening the doors for easy experimental and deployment-specific TCP
option use.
- Reorganize TCP congestion control (CC) initialization to simplify
life of TCP CC implemented in BPF.
- Add support for shipping BPF programs with the kernel and loading
them early on boot via the User Mode Driver mechanism, hence reusing
all the user space infra we have.
- Support sleepable BPF programs, initially targeting LSM and tracing.
- Add bpf_d_path() helper for returning full path for given 'struct
path'.
- Make bpf_tail_call compatible with bpf-to-bpf calls.
- Allow BPF programs to call map_update_elem on sockmaps.
- Add BPF Type Format (BTF) support for type and enum discovery, as
well as support for using BTF within the kernel itself (current use
is for pretty printing structures).
- Support listing and getting information about bpf_links via the bpf
syscall.
- Enhance kernel interfaces around NIC firmware update. Allow
specifying overwrite mask to control if settings etc. are reset
during update; report expected max time operation may take to users;
support firmware activation without machine reboot incl. limits of
how much impact reset may have (e.g. dropping link or not).
- Extend ethtool configuration interface to report IEEE-standard
counters, to limit the need for per-vendor logic in user space.
- Adopt or extend devlink use for debug, monitoring, fw update in many
drivers (dsa loop, ice, ionic, sja1105, qed, mlxsw, mv88e6xxx,
dpaa2-eth).
- In mlxsw expose critical and emergency SFP module temperature alarms.
Refactor port buffer handling to make the defaults more suitable and
support setting these values explicitly via the DCBNL interface.
- Add XDP support for Intel's igb driver.
- Support offloading TC flower classification and filtering rules to
mscc_ocelot switches.
- Add PTP support for Marvell Octeontx2 and PP2.2 hardware, as well as
fixed interval period pulse generator and one-step timestamping in
dpaa-eth.
- Add support for various auth offloads in WiFi APs, e.g. SAE (WPA3)
offload.
- Add Lynx PHY/PCS MDIO module, and convert various drivers which have
this HW to use it. Convert mvpp2 to split PCS.
- Support Marvell Prestera 98DX3255 24-port switch ASICs, as well as
7-port Mediatek MT7531 IP.
- Add initial support for QCA6390 and IPQ6018 in ath11k WiFi driver,
and wcn3680 support in wcn36xx.
- Improve performance for packets which don't require much offloads on
recent Mellanox NICs by 20% by making multiple packets share a
descriptor entry.
- Move chelsio inline crypto drivers (for TLS and IPsec) from the
crypto subtree to drivers/net. Move MDIO drivers out of the phy
directory.
- Clean up a lot of W=1 warnings, reportedly the actively developed
subsections of networking drivers should now build W=1 warning free.
- Make sure drivers don't use in_interrupt() to dynamically adapt their
code. Convert tasklets to use new tasklet_setup API (sadly this
conversion is not yet complete).
* tag 'net-next-5.10' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (2583 commits)
Revert "bpfilter: Fix build error with CONFIG_BPFILTER_UMH"
net, sockmap: Don't call bpf_prog_put() on NULL pointer
bpf, selftest: Fix flaky tcp_hdr_options test when adding addr to lo
bpf, sockmap: Add locking annotations to iterator
netfilter: nftables: allow re-computing sctp CRC-32C in 'payload' statements
net: fix pos incrementment in ipv6_route_seq_next
net/smc: fix invalid return code in smcd_new_buf_create()
net/smc: fix valid DMBE buffer sizes
net/smc: fix use-after-free of delayed events
bpfilter: Fix build error with CONFIG_BPFILTER_UMH
cxgb4/ch_ipsec: Replace the module name to ch_ipsec from chcr
net: sched: Fix suspicious RCU usage while accessing tcf_tunnel_info
bpf: Fix register equivalence tracking.
rxrpc: Fix loss of final ack on shutdown
rxrpc: Fix bundle counting for exclusive connections
netfilter: restore NF_INET_NUMHOOKS
ibmveth: Identify ingress large send packets.
ibmveth: Switch order of ibmveth_helper calls.
cxgb4: handle 4-tuple PEDIT to NAT mode translation
selftests: Add VRF route leaking tests
...
Diffstat (limited to 'drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c')
-rw-r--r-- | drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c | 911 |
1 files changed, 911 insertions, 0 deletions
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c new file mode 100644 index 000000000000..947f346bdc2d --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c @@ -0,0 +1,911 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +// Copyright (c) 2020 Mellanox Technologies. + +#include <linux/mlx5/driver.h> +#include <linux/mlx5/mlx5_ifc.h> +#include <linux/mlx5/fs.h> + +#include "lib/fs_chains.h" +#include "en/mapping.h" +#include "mlx5_core.h" +#include "fs_core.h" +#include "eswitch.h" +#include "en.h" +#include "en_tc.h" + +#define chains_lock(chains) ((chains)->lock) +#define chains_ht(chains) ((chains)->chains_ht) +#define chains_mapping(chains) ((chains)->chains_mapping) +#define prios_ht(chains) ((chains)->prios_ht) +#define ft_pool_left(chains) ((chains)->ft_left) +#define tc_default_ft(chains) ((chains)->tc_default_ft) +#define tc_end_ft(chains) ((chains)->tc_end_ft) +#define ns_to_chains_fs_prio(ns) ((ns) == MLX5_FLOW_NAMESPACE_FDB ? \ + FDB_TC_OFFLOAD : MLX5E_TC_PRIO) + +/* Firmware currently has 4 pool of 4 sizes that it supports (FT_POOLS), + * and a virtual memory region of 16M (MLX5_FT_SIZE), this region is duplicated + * for each flow table pool. We can allocate up to 16M of each pool, + * and we keep track of how much we used via get_next_avail_sz_from_pool. + * Firmware doesn't report any of this for now. + * ESW_POOL is expected to be sorted from large to small and match firmware + * pools. + */ +#define FT_SIZE (16 * 1024 * 1024) +static const unsigned int FT_POOLS[] = { 4 * 1024 * 1024, + 1 * 1024 * 1024, + 64 * 1024, + 128 }; +#define FT_TBL_SZ (64 * 1024) + +struct mlx5_fs_chains { + struct mlx5_core_dev *dev; + + struct rhashtable chains_ht; + struct rhashtable prios_ht; + /* Protects above chains_ht and prios_ht */ + struct mutex lock; + + struct mlx5_flow_table *tc_default_ft; + struct mlx5_flow_table *tc_end_ft; + struct mapping_ctx *chains_mapping; + + enum mlx5_flow_namespace_type ns; + u32 group_num; + u32 flags; + + int ft_left[ARRAY_SIZE(FT_POOLS)]; +}; + +struct fs_chain { + struct rhash_head node; + + u32 chain; + + int ref; + int id; + + struct mlx5_fs_chains *chains; + struct list_head prios_list; + struct mlx5_flow_handle *restore_rule; + struct mlx5_modify_hdr *miss_modify_hdr; +}; + +struct prio_key { + u32 chain; + u32 prio; + u32 level; +}; + +struct prio { + struct rhash_head node; + struct list_head list; + + struct prio_key key; + + int ref; + + struct fs_chain *chain; + struct mlx5_flow_table *ft; + struct mlx5_flow_table *next_ft; + struct mlx5_flow_group *miss_group; + struct mlx5_flow_handle *miss_rule; +}; + +static const struct rhashtable_params chain_params = { + .head_offset = offsetof(struct fs_chain, node), + .key_offset = offsetof(struct fs_chain, chain), + .key_len = sizeof_field(struct fs_chain, chain), + .automatic_shrinking = true, +}; + +static const struct rhashtable_params prio_params = { + .head_offset = offsetof(struct prio, node), + .key_offset = offsetof(struct prio, key), + .key_len = sizeof_field(struct prio, key), + .automatic_shrinking = true, +}; + +bool mlx5_chains_prios_supported(struct mlx5_fs_chains *chains) +{ + return chains->flags & MLX5_CHAINS_AND_PRIOS_SUPPORTED; +} + +static bool mlx5_chains_ignore_flow_level_supported(struct mlx5_fs_chains *chains) +{ + return chains->flags & MLX5_CHAINS_IGNORE_FLOW_LEVEL_SUPPORTED; +} + +bool mlx5_chains_backwards_supported(struct mlx5_fs_chains *chains) +{ + return mlx5_chains_prios_supported(chains) && + mlx5_chains_ignore_flow_level_supported(chains); +} + +u32 mlx5_chains_get_chain_range(struct mlx5_fs_chains *chains) +{ + if (!mlx5_chains_prios_supported(chains)) + return 1; + + if (mlx5_chains_ignore_flow_level_supported(chains)) + return UINT_MAX - 1; + + /* We should get here only for eswitch case */ + return FDB_TC_MAX_CHAIN; +} + +u32 mlx5_chains_get_nf_ft_chain(struct mlx5_fs_chains *chains) +{ + return mlx5_chains_get_chain_range(chains) + 1; +} + +u32 mlx5_chains_get_prio_range(struct mlx5_fs_chains *chains) +{ + if (!mlx5_chains_prios_supported(chains)) + return 1; + + if (mlx5_chains_ignore_flow_level_supported(chains)) + return UINT_MAX; + + /* We should get here only for eswitch case */ + return FDB_TC_MAX_PRIO; +} + +static unsigned int mlx5_chains_get_level_range(struct mlx5_fs_chains *chains) +{ + if (mlx5_chains_ignore_flow_level_supported(chains)) + return UINT_MAX; + + /* Same value for FDB and NIC RX tables */ + return FDB_TC_LEVELS_PER_PRIO; +} + +void +mlx5_chains_set_end_ft(struct mlx5_fs_chains *chains, + struct mlx5_flow_table *ft) +{ + tc_end_ft(chains) = ft; +} + +#define POOL_NEXT_SIZE 0 +static int +mlx5_chains_get_avail_sz_from_pool(struct mlx5_fs_chains *chains, + int desired_size) +{ + int i, found_i = -1; + + for (i = ARRAY_SIZE(FT_POOLS) - 1; i >= 0; i--) { + if (ft_pool_left(chains)[i] && FT_POOLS[i] > desired_size) { + found_i = i; + if (desired_size != POOL_NEXT_SIZE) + break; + } + } + + if (found_i != -1) { + --ft_pool_left(chains)[found_i]; + return FT_POOLS[found_i]; + } + + return 0; +} + +static void +mlx5_chains_put_sz_to_pool(struct mlx5_fs_chains *chains, int sz) +{ + int i; + + for (i = ARRAY_SIZE(FT_POOLS) - 1; i >= 0; i--) { + if (sz == FT_POOLS[i]) { + ++ft_pool_left(chains)[i]; + return; + } + } + + WARN_ONCE(1, "Couldn't find size %d in flow table size pool", sz); +} + +static void +mlx5_chains_init_sz_pool(struct mlx5_fs_chains *chains, u32 ft_max) +{ + int i; + + for (i = ARRAY_SIZE(FT_POOLS) - 1; i >= 0; i--) + ft_pool_left(chains)[i] = + FT_POOLS[i] <= ft_max ? FT_SIZE / FT_POOLS[i] : 0; +} + +static struct mlx5_flow_table * +mlx5_chains_create_table(struct mlx5_fs_chains *chains, + u32 chain, u32 prio, u32 level) +{ + struct mlx5_flow_table_attr ft_attr = {}; + struct mlx5_flow_namespace *ns; + struct mlx5_flow_table *ft; + int sz; + + if (chains->flags & MLX5_CHAINS_FT_TUNNEL_SUPPORTED) + ft_attr.flags |= (MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT | + MLX5_FLOW_TABLE_TUNNEL_EN_DECAP); + + sz = (chain == mlx5_chains_get_nf_ft_chain(chains)) ? + mlx5_chains_get_avail_sz_from_pool(chains, FT_TBL_SZ) : + mlx5_chains_get_avail_sz_from_pool(chains, POOL_NEXT_SIZE); + if (!sz) + return ERR_PTR(-ENOSPC); + ft_attr.max_fte = sz; + + /* We use tc_default_ft(chains) as the table's next_ft till + * ignore_flow_level is allowed on FT creation and not just for FTEs. + * Instead caller should add an explicit miss rule if needed. + */ + ft_attr.next_ft = tc_default_ft(chains); + + /* The root table(chain 0, prio 1, level 0) is required to be + * connected to the previous fs_core managed prio. + * We always create it, as a managed table, in order to align with + * fs_core logic. + */ + if (!mlx5_chains_ignore_flow_level_supported(chains) || + (chain == 0 && prio == 1 && level == 0)) { + ft_attr.level = level; + ft_attr.prio = prio - 1; + ns = (chains->ns == MLX5_FLOW_NAMESPACE_FDB) ? + mlx5_get_fdb_sub_ns(chains->dev, chain) : + mlx5_get_flow_namespace(chains->dev, chains->ns); + } else { + ft_attr.flags |= MLX5_FLOW_TABLE_UNMANAGED; + ft_attr.prio = ns_to_chains_fs_prio(chains->ns); + /* Firmware doesn't allow us to create another level 0 table, + * so we create all unmanaged tables as level 1. + * + * To connect them, we use explicit miss rules with + * ignore_flow_level. Caller is responsible to create + * these rules (if needed). + */ + ft_attr.level = 1; + ns = mlx5_get_flow_namespace(chains->dev, chains->ns); + } + + ft_attr.autogroup.num_reserved_entries = 2; + ft_attr.autogroup.max_num_groups = chains->group_num; + ft = mlx5_create_auto_grouped_flow_table(ns, &ft_attr); + if (IS_ERR(ft)) { + mlx5_core_warn(chains->dev, "Failed to create chains table err %d (chain: %d, prio: %d, level: %d, size: %d)\n", + (int)PTR_ERR(ft), chain, prio, level, sz); + mlx5_chains_put_sz_to_pool(chains, sz); + return ft; + } + + return ft; +} + +static void +mlx5_chains_destroy_table(struct mlx5_fs_chains *chains, + struct mlx5_flow_table *ft) +{ + mlx5_chains_put_sz_to_pool(chains, ft->max_fte); + mlx5_destroy_flow_table(ft); +} + +static int +create_chain_restore(struct fs_chain *chain) +{ + struct mlx5_eswitch *esw = chain->chains->dev->priv.eswitch; + char modact[MLX5_UN_SZ_BYTES(set_add_copy_action_in_auto)]; + struct mlx5_fs_chains *chains = chain->chains; + enum mlx5e_tc_attr_to_reg chain_to_reg; + struct mlx5_modify_hdr *mod_hdr; + u32 index; + int err; + + if (chain->chain == mlx5_chains_get_nf_ft_chain(chains) || + !mlx5_chains_prios_supported(chains)) + return 0; + + err = mapping_add(chains_mapping(chains), &chain->chain, &index); + if (err) + return err; + if (index == MLX5_FS_DEFAULT_FLOW_TAG) { + /* we got the special default flow tag id, so we won't know + * if we actually marked the packet with the restore rule + * we create. + * + * This case isn't possible with MLX5_FS_DEFAULT_FLOW_TAG = 0. + */ + err = mapping_add(chains_mapping(chains), + &chain->chain, &index); + mapping_remove(chains_mapping(chains), + MLX5_FS_DEFAULT_FLOW_TAG); + if (err) + return err; + } + + chain->id = index; + + if (chains->ns == MLX5_FLOW_NAMESPACE_FDB) { + chain_to_reg = CHAIN_TO_REG; + chain->restore_rule = esw_add_restore_rule(esw, chain->id); + if (IS_ERR(chain->restore_rule)) { + err = PTR_ERR(chain->restore_rule); + goto err_rule; + } + } else if (chains->ns == MLX5_FLOW_NAMESPACE_KERNEL) { + /* For NIC RX we don't need a restore rule + * since we write the metadata to reg_b + * that is passed to SW directly. + */ + chain_to_reg = NIC_CHAIN_TO_REG; + } else { + err = -EINVAL; + goto err_rule; + } + + MLX5_SET(set_action_in, modact, action_type, MLX5_ACTION_TYPE_SET); + MLX5_SET(set_action_in, modact, field, + mlx5e_tc_attr_to_reg_mappings[chain_to_reg].mfield); + MLX5_SET(set_action_in, modact, offset, + mlx5e_tc_attr_to_reg_mappings[chain_to_reg].moffset * 8); + MLX5_SET(set_action_in, modact, length, + mlx5e_tc_attr_to_reg_mappings[chain_to_reg].mlen * 8); + MLX5_SET(set_action_in, modact, data, chain->id); + mod_hdr = mlx5_modify_header_alloc(chains->dev, chains->ns, + 1, modact); + if (IS_ERR(mod_hdr)) { + err = PTR_ERR(mod_hdr); + goto err_mod_hdr; + } + chain->miss_modify_hdr = mod_hdr; + + return 0; + +err_mod_hdr: + if (!IS_ERR_OR_NULL(chain->restore_rule)) + mlx5_del_flow_rules(chain->restore_rule); +err_rule: + /* Datapath can't find this mapping, so we can safely remove it */ + mapping_remove(chains_mapping(chains), chain->id); + return err; +} + +static void destroy_chain_restore(struct fs_chain *chain) +{ + struct mlx5_fs_chains *chains = chain->chains; + + if (!chain->miss_modify_hdr) + return; + + if (chain->restore_rule) + mlx5_del_flow_rules(chain->restore_rule); + + mlx5_modify_header_dealloc(chains->dev, chain->miss_modify_hdr); + mapping_remove(chains_mapping(chains), chain->id); +} + +static struct fs_chain * +mlx5_chains_create_chain(struct mlx5_fs_chains *chains, u32 chain) +{ + struct fs_chain *chain_s = NULL; + int err; + + chain_s = kvzalloc(sizeof(*chain_s), GFP_KERNEL); + if (!chain_s) + return ERR_PTR(-ENOMEM); + + chain_s->chains = chains; + chain_s->chain = chain; + INIT_LIST_HEAD(&chain_s->prios_list); + + err = create_chain_restore(chain_s); + if (err) + goto err_restore; + + err = rhashtable_insert_fast(&chains_ht(chains), &chain_s->node, + chain_params); + if (err) + goto err_insert; + + return chain_s; + +err_insert: + destroy_chain_restore(chain_s); +err_restore: + kvfree(chain_s); + return ERR_PTR(err); +} + +static void +mlx5_chains_destroy_chain(struct fs_chain *chain) +{ + struct mlx5_fs_chains *chains = chain->chains; + + rhashtable_remove_fast(&chains_ht(chains), &chain->node, + chain_params); + + destroy_chain_restore(chain); + kvfree(chain); +} + +static struct fs_chain * +mlx5_chains_get_chain(struct mlx5_fs_chains *chains, u32 chain) +{ + struct fs_chain *chain_s; + + chain_s = rhashtable_lookup_fast(&chains_ht(chains), &chain, + chain_params); + if (!chain_s) { + chain_s = mlx5_chains_create_chain(chains, chain); + if (IS_ERR(chain_s)) + return chain_s; + } + + chain_s->ref++; + + return chain_s; +} + +static struct mlx5_flow_handle * +mlx5_chains_add_miss_rule(struct fs_chain *chain, + struct mlx5_flow_table *ft, + struct mlx5_flow_table *next_ft) +{ + struct mlx5_fs_chains *chains = chain->chains; + struct mlx5_flow_destination dest = {}; + struct mlx5_flow_act act = {}; + + act.flags = FLOW_ACT_NO_APPEND; + if (mlx5_chains_ignore_flow_level_supported(chain->chains)) + act.flags |= FLOW_ACT_IGNORE_FLOW_LEVEL; + + act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; + dest.ft = next_ft; + + if (next_ft == tc_end_ft(chains) && + chain->chain != mlx5_chains_get_nf_ft_chain(chains) && + mlx5_chains_prios_supported(chains)) { + act.modify_hdr = chain->miss_modify_hdr; + act.action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; + } + + return mlx5_add_flow_rules(ft, NULL, &act, &dest, 1); +} + +static int +mlx5_chains_update_prio_prevs(struct prio *prio, + struct mlx5_flow_table *next_ft) +{ + struct mlx5_flow_handle *miss_rules[FDB_TC_LEVELS_PER_PRIO + 1] = {}; + struct fs_chain *chain = prio->chain; + struct prio *pos; + int n = 0, err; + + if (prio->key.level) + return 0; + + /* Iterate in reverse order until reaching the level 0 rule of + * the previous priority, adding all the miss rules first, so we can + * revert them if any of them fails. + */ + pos = prio; + list_for_each_entry_continue_reverse(pos, + &chain->prios_list, + list) { + miss_rules[n] = mlx5_chains_add_miss_rule(chain, + pos->ft, + next_ft); + if (IS_ERR(miss_rules[n])) { + err = PTR_ERR(miss_rules[n]); + goto err_prev_rule; + } + + n++; + if (!pos->key.level) + break; + } + + /* Success, delete old miss rules, and update the pointers. */ + n = 0; + pos = prio; + list_for_each_entry_continue_reverse(pos, + &chain->prios_list, + list) { + mlx5_del_flow_rules(pos->miss_rule); + + pos->miss_rule = miss_rules[n]; + pos->next_ft = next_ft; + + n++; + if (!pos->key.level) + break; + } + + return 0; + +err_prev_rule: + while (--n >= 0) + mlx5_del_flow_rules(miss_rules[n]); + + return err; +} + +static void +mlx5_chains_put_chain(struct fs_chain *chain) +{ + if (--chain->ref == 0) + mlx5_chains_destroy_chain(chain); +} + +static struct prio * +mlx5_chains_create_prio(struct mlx5_fs_chains *chains, + u32 chain, u32 prio, u32 level) +{ + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct mlx5_flow_handle *miss_rule = NULL; + struct mlx5_flow_group *miss_group; + struct mlx5_flow_table *next_ft; + struct mlx5_flow_table *ft; + struct prio *prio_s = NULL; + struct fs_chain *chain_s; + struct list_head *pos; + u32 *flow_group_in; + int err; + + chain_s = mlx5_chains_get_chain(chains, chain); + if (IS_ERR(chain_s)) + return ERR_CAST(chain_s); + + prio_s = kvzalloc(sizeof(*prio_s), GFP_KERNEL); + flow_group_in = kvzalloc(inlen, GFP_KERNEL); + if (!prio_s || !flow_group_in) { + err = -ENOMEM; + goto err_alloc; + } + + /* Chain's prio list is sorted by prio and level. + * And all levels of some prio point to the next prio's level 0. + * Example list (prio, level): + * (3,0)->(3,1)->(5,0)->(5,1)->(6,1)->(7,0) + * In hardware, we will we have the following pointers: + * (3,0) -> (5,0) -> (7,0) -> Slow path + * (3,1) -> (5,0) + * (5,1) -> (7,0) + * (6,1) -> (7,0) + */ + + /* Default miss for each chain: */ + next_ft = (chain == mlx5_chains_get_nf_ft_chain(chains)) ? + tc_default_ft(chains) : + tc_end_ft(chains); + list_for_each(pos, &chain_s->prios_list) { + struct prio *p = list_entry(pos, struct prio, list); + + /* exit on first pos that is larger */ + if (prio < p->key.prio || (prio == p->key.prio && + level < p->key.level)) { + /* Get next level 0 table */ + next_ft = p->key.level == 0 ? p->ft : p->next_ft; + break; + } + } + + ft = mlx5_chains_create_table(chains, chain, prio, level); + if (IS_ERR(ft)) { + err = PTR_ERR(ft); + goto err_create; + } + + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, + ft->max_fte - 2); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, + ft->max_fte - 1); + miss_group = mlx5_create_flow_group(ft, flow_group_in); + if (IS_ERR(miss_group)) { + err = PTR_ERR(miss_group); + goto err_group; + } + + /* Add miss rule to next_ft */ + miss_rule = mlx5_chains_add_miss_rule(chain_s, ft, next_ft); + if (IS_ERR(miss_rule)) { + err = PTR_ERR(miss_rule); + goto err_miss_rule; + } + + prio_s->miss_group = miss_group; + prio_s->miss_rule = miss_rule; + prio_s->next_ft = next_ft; + prio_s->chain = chain_s; + prio_s->key.chain = chain; + prio_s->key.prio = prio; + prio_s->key.level = level; + prio_s->ft = ft; + + err = rhashtable_insert_fast(&prios_ht(chains), &prio_s->node, + prio_params); + if (err) + goto err_insert; + + list_add(&prio_s->list, pos->prev); + + /* Table is ready, connect it */ + err = mlx5_chains_update_prio_prevs(prio_s, ft); + if (err) + goto err_update; + + kvfree(flow_group_in); + return prio_s; + +err_update: + list_del(&prio_s->list); + rhashtable_remove_fast(&prios_ht(chains), &prio_s->node, + prio_params); +err_insert: + mlx5_del_flow_rules(miss_rule); +err_miss_rule: + mlx5_destroy_flow_group(miss_group); +err_group: + mlx5_chains_destroy_table(chains, ft); +err_create: +err_alloc: + kvfree(prio_s); + kvfree(flow_group_in); + mlx5_chains_put_chain(chain_s); + return ERR_PTR(err); +} + +static void +mlx5_chains_destroy_prio(struct mlx5_fs_chains *chains, + struct prio *prio) +{ + struct fs_chain *chain = prio->chain; + + WARN_ON(mlx5_chains_update_prio_prevs(prio, + prio->next_ft)); + + list_del(&prio->list); + rhashtable_remove_fast(&prios_ht(chains), &prio->node, + prio_params); + mlx5_del_flow_rules(prio->miss_rule); + mlx5_destroy_flow_group(prio->miss_group); + mlx5_chains_destroy_table(chains, prio->ft); + mlx5_chains_put_chain(chain); + kvfree(prio); +} + +struct mlx5_flow_table * +mlx5_chains_get_table(struct mlx5_fs_chains *chains, u32 chain, u32 prio, + u32 level) +{ + struct mlx5_flow_table *prev_fts; + struct prio *prio_s; + struct prio_key key; + int l = 0; + + if ((chain > mlx5_chains_get_chain_range(chains) && + chain != mlx5_chains_get_nf_ft_chain(chains)) || + prio > mlx5_chains_get_prio_range(chains) || + level > mlx5_chains_get_level_range(chains)) + return ERR_PTR(-EOPNOTSUPP); + + /* create earlier levels for correct fs_core lookup when + * connecting tables. + */ + for (l = 0; l < level; l++) { + prev_fts = mlx5_chains_get_table(chains, chain, prio, l); + if (IS_ERR(prev_fts)) { + prio_s = ERR_CAST(prev_fts); + goto err_get_prevs; + } + } + + key.chain = chain; + key.prio = prio; + key.level = level; + + mutex_lock(&chains_lock(chains)); + prio_s = rhashtable_lookup_fast(&prios_ht(chains), &key, + prio_params); + if (!prio_s) { + prio_s = mlx5_chains_create_prio(chains, chain, + prio, level); + if (IS_ERR(prio_s)) + goto err_create_prio; + } + + ++prio_s->ref; + mutex_unlock(&chains_lock(chains)); + + return prio_s->ft; + +err_create_prio: + mutex_unlock(&chains_lock(chains)); +err_get_prevs: + while (--l >= 0) + mlx5_chains_put_table(chains, chain, prio, l); + return ERR_CAST(prio_s); +} + +void +mlx5_chains_put_table(struct mlx5_fs_chains *chains, u32 chain, u32 prio, + u32 level) +{ + struct prio *prio_s; + struct prio_key key; + + key.chain = chain; + key.prio = prio; + key.level = level; + + mutex_lock(&chains_lock(chains)); + prio_s = rhashtable_lookup_fast(&prios_ht(chains), &key, + prio_params); + if (!prio_s) + goto err_get_prio; + + if (--prio_s->ref == 0) + mlx5_chains_destroy_prio(chains, prio_s); + mutex_unlock(&chains_lock(chains)); + + while (level-- > 0) + mlx5_chains_put_table(chains, chain, prio, level); + + return; + +err_get_prio: + mutex_unlock(&chains_lock(chains)); + WARN_ONCE(1, + "Couldn't find table: (chain: %d prio: %d level: %d)", + chain, prio, level); +} + +struct mlx5_flow_table * +mlx5_chains_get_tc_end_ft(struct mlx5_fs_chains *chains) +{ + return tc_end_ft(chains); +} + +struct mlx5_flow_table * +mlx5_chains_create_global_table(struct mlx5_fs_chains *chains) +{ + u32 chain, prio, level; + int err; + + if (!mlx5_chains_ignore_flow_level_supported(chains)) { + err = -EOPNOTSUPP; + + mlx5_core_warn(chains->dev, + "Couldn't create global flow table, ignore_flow_level not supported."); + goto err_ignore; + } + + chain = mlx5_chains_get_chain_range(chains), + prio = mlx5_chains_get_prio_range(chains); + level = mlx5_chains_get_level_range(chains); + + return mlx5_chains_create_table(chains, chain, prio, level); + +err_ignore: + return ERR_PTR(err); +} + +void +mlx5_chains_destroy_global_table(struct mlx5_fs_chains *chains, + struct mlx5_flow_table *ft) +{ + mlx5_chains_destroy_table(chains, ft); +} + +static struct mlx5_fs_chains * +mlx5_chains_init(struct mlx5_core_dev *dev, struct mlx5_chains_attr *attr) +{ + struct mlx5_fs_chains *chains_priv; + struct mapping_ctx *mapping; + u32 max_flow_counter; + int err; + + chains_priv = kzalloc(sizeof(*chains_priv), GFP_KERNEL); + if (!chains_priv) + return ERR_PTR(-ENOMEM); + + max_flow_counter = (MLX5_CAP_GEN(dev, max_flow_counter_31_16) << 16) | + MLX5_CAP_GEN(dev, max_flow_counter_15_0); + + mlx5_core_dbg(dev, + "Init flow table chains, max counters(%d), groups(%d), max flow table size(%d)\n", + max_flow_counter, attr->max_grp_num, attr->max_ft_sz); + + chains_priv->dev = dev; + chains_priv->flags = attr->flags; + chains_priv->ns = attr->ns; + chains_priv->group_num = attr->max_grp_num; + tc_default_ft(chains_priv) = tc_end_ft(chains_priv) = attr->default_ft; + + mlx5_core_info(dev, "Supported tc offload range - chains: %u, prios: %u\n", + mlx5_chains_get_chain_range(chains_priv), + mlx5_chains_get_prio_range(chains_priv)); + + mlx5_chains_init_sz_pool(chains_priv, attr->max_ft_sz); + + err = rhashtable_init(&chains_ht(chains_priv), &chain_params); + if (err) + goto init_chains_ht_err; + + err = rhashtable_init(&prios_ht(chains_priv), &prio_params); + if (err) + goto init_prios_ht_err; + + mapping = mapping_create(sizeof(u32), attr->max_restore_tag, + true); + if (IS_ERR(mapping)) { + err = PTR_ERR(mapping); + goto mapping_err; + } + chains_mapping(chains_priv) = mapping; + + mutex_init(&chains_lock(chains_priv)); + + return chains_priv; + +mapping_err: + rhashtable_destroy(&prios_ht(chains_priv)); +init_prios_ht_err: + rhashtable_destroy(&chains_ht(chains_priv)); +init_chains_ht_err: + kfree(chains_priv); + return ERR_PTR(err); +} + +static void +mlx5_chains_cleanup(struct mlx5_fs_chains *chains) +{ + mutex_destroy(&chains_lock(chains)); + mapping_destroy(chains_mapping(chains)); + rhashtable_destroy(&prios_ht(chains)); + rhashtable_destroy(&chains_ht(chains)); + + kfree(chains); +} + +struct mlx5_fs_chains * +mlx5_chains_create(struct mlx5_core_dev *dev, struct mlx5_chains_attr *attr) +{ + struct mlx5_fs_chains *chains; + + chains = mlx5_chains_init(dev, attr); + + return chains; +} + +void +mlx5_chains_destroy(struct mlx5_fs_chains *chains) +{ + mlx5_chains_cleanup(chains); +} + +int +mlx5_chains_get_chain_mapping(struct mlx5_fs_chains *chains, u32 chain, + u32 *chain_mapping) +{ + return mapping_add(chains_mapping(chains), &chain, chain_mapping); +} + +int +mlx5_chains_put_chain_mapping(struct mlx5_fs_chains *chains, u32 chain_mapping) +{ + return mapping_remove(chains_mapping(chains), chain_mapping); +} + +int mlx5_get_chain_for_tag(struct mlx5_fs_chains *chains, u32 tag, + u32 *chain) +{ + int err; + + err = mapping_find(chains_mapping(chains), tag, chain); + if (err) { + mlx5_core_warn(chains->dev, "Can't find chain for tag: %d\n", tag); + return -ENOENT; + } + + return 0; +} |