From 49df08ddbdb7211561a2b1e1ca66fa7b2ea20fb2 Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Tue, 1 Oct 2019 12:38:21 -0300 Subject: RDMA/mlx5: Add missing synchronize_srcu() for MW cases [ Upstream commit 0417791536ae1e28d7f0418f1d20048ec4d3c6cf ] While MR uses live as the SRCU 'update', the MW case uses the xarray directly, xa_erase() causes the MW to become inaccessible to the pagefault thread. Thus whenever a MW is removed from the xarray we must synchronize_srcu() before freeing it. This must be done before freeing the mkey as re-use of the mkey while the pagefault thread is using the stale mkey is undesirable. Add the missing synchronizes to MW and DEVX indirect mkey and delete the bogus protection against double destroy in mlx5_core_destroy_mkey() Fixes: 534fd7aac56a ("IB/mlx5: Manage indirection mkey upon DEVX flow for ODP") Fixes: 6aec21f6a832 ("IB/mlx5: Page faults handling infrastructure") Link: https://lore.kernel.org/r/20191001153821.23621-7-jgg@ziepe.ca Reviewed-by: Artemy Kovalyov Signed-off-by: Jason Gunthorpe Signed-off-by: Sasha Levin --- drivers/net/ethernet/mellanox/mlx5/core/mr.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) (limited to 'drivers/net/ethernet/mellanox') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mr.c b/drivers/net/ethernet/mellanox/mlx5/core/mr.c index 9231b39d18b2..c501bf2a0252 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mr.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/mr.c @@ -112,17 +112,11 @@ int mlx5_core_destroy_mkey(struct mlx5_core_dev *dev, u32 out[MLX5_ST_SZ_DW(destroy_mkey_out)] = {0}; u32 in[MLX5_ST_SZ_DW(destroy_mkey_in)] = {0}; struct xarray *mkeys = &dev->priv.mkey_table; - struct mlx5_core_mkey *deleted_mkey; unsigned long flags; xa_lock_irqsave(mkeys, flags); - deleted_mkey = __xa_erase(mkeys, mlx5_base_mkey(mkey->key)); + __xa_erase(mkeys, mlx5_base_mkey(mkey->key)); xa_unlock_irqrestore(mkeys, flags); - if (!deleted_mkey) { - mlx5_core_dbg(dev, "failed xarray delete of mkey 0x%x\n", - mlx5_base_mkey(mkey->key)); - return -ENOENT; - } MLX5_SET(destroy_mkey_in, in, opcode, MLX5_CMD_OP_DESTROY_MKEY); MLX5_SET(destroy_mkey_in, in, mkey_index, mlx5_mkey_to_idx(mkey->key)); -- cgit v1.2.3 From 02ccce8bb95c5a9c0fcf886790dd98d910ecb064 Mon Sep 17 00:00:00 2001 From: Eran Ben Elisha Date: Sun, 27 Oct 2019 16:39:15 +0200 Subject: net/mlx4_core: Dynamically set guaranteed amount of counters per VF [ Upstream commit e19868efea0c103f23b4b7e986fd0a703822111f ] Prior to this patch, the amount of counters guaranteed per VF in the resource tracker was MLX4_VF_COUNTERS_PER_PORT * MLX4_MAX_PORTS. It was set regardless if the VF was single or dual port. This caused several VFs to have no guaranteed counters although the system could satisfy their request. The fix is to dynamically guarantee counters, based on each VF specification. Fixes: 9de92c60beaa ("net/mlx4_core: Adjust counter grant policy in the resource tracker") Signed-off-by: Eran Ben Elisha Signed-off-by: Jack Morgenstein Signed-off-by: Tariq Toukan Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- .../net/ethernet/mellanox/mlx4/resource_tracker.c | 42 +++++++++++++--------- 1 file changed, 26 insertions(+), 16 deletions(-) (limited to 'drivers/net/ethernet/mellanox') diff --git a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c index 4356f3a58002..1187ef1375e2 100644 --- a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c +++ b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c @@ -471,12 +471,31 @@ void mlx4_init_quotas(struct mlx4_dev *dev) priv->mfunc.master.res_tracker.res_alloc[RES_MPT].quota[pf]; } -static int get_max_gauranteed_vfs_counter(struct mlx4_dev *dev) +static int +mlx4_calc_res_counter_guaranteed(struct mlx4_dev *dev, + struct resource_allocator *res_alloc, + int vf) { - /* reduce the sink counter */ - return (dev->caps.max_counters - 1 - - (MLX4_PF_COUNTERS_PER_PORT * MLX4_MAX_PORTS)) - / MLX4_MAX_PORTS; + struct mlx4_active_ports actv_ports; + int ports, counters_guaranteed; + + /* For master, only allocate according to the number of phys ports */ + if (vf == mlx4_master_func_num(dev)) + return MLX4_PF_COUNTERS_PER_PORT * dev->caps.num_ports; + + /* calculate real number of ports for the VF */ + actv_ports = mlx4_get_active_ports(dev, vf); + ports = bitmap_weight(actv_ports.ports, dev->caps.num_ports); + counters_guaranteed = ports * MLX4_VF_COUNTERS_PER_PORT; + + /* If we do not have enough counters for this VF, do not + * allocate any for it. '-1' to reduce the sink counter. + */ + if ((res_alloc->res_reserved + counters_guaranteed) > + (dev->caps.max_counters - 1)) + return 0; + + return counters_guaranteed; } int mlx4_init_resource_tracker(struct mlx4_dev *dev) @@ -484,7 +503,6 @@ int mlx4_init_resource_tracker(struct mlx4_dev *dev) struct mlx4_priv *priv = mlx4_priv(dev); int i, j; int t; - int max_vfs_guarantee_counter = get_max_gauranteed_vfs_counter(dev); priv->mfunc.master.res_tracker.slave_list = kcalloc(dev->num_slaves, sizeof(struct slave_list), @@ -603,16 +621,8 @@ int mlx4_init_resource_tracker(struct mlx4_dev *dev) break; case RES_COUNTER: res_alloc->quota[t] = dev->caps.max_counters; - if (t == mlx4_master_func_num(dev)) - res_alloc->guaranteed[t] = - MLX4_PF_COUNTERS_PER_PORT * - MLX4_MAX_PORTS; - else if (t <= max_vfs_guarantee_counter) - res_alloc->guaranteed[t] = - MLX4_VF_COUNTERS_PER_PORT * - MLX4_MAX_PORTS; - else - res_alloc->guaranteed[t] = 0; + res_alloc->guaranteed[t] = + mlx4_calc_res_counter_guaranteed(dev, res_alloc, t); break; default: break; -- cgit v1.2.3 From 11686ceb20a8996baf750f13b49f3746a6596e2c Mon Sep 17 00:00:00 2001 From: Dmytro Linkin Date: Wed, 4 Sep 2019 12:32:49 +0000 Subject: net/mlx5e: Determine source port properly for vlan push action [ Upstream commit d5dbcc4e87bc8444bd2f1ca4b8f787e1e5677ec2 ] Termination tables are used for vlan push actions on uplink ports. To support RoCE dual port the source port value was placed in a register. Fix the code to use an API method returning the source port according to the FW capabilities. Fixes: 10caabdaad5a ("net/mlx5e: Use termination table for VLAN push actions") Signed-off-by: Dmytro Linkin Reviewed-by: Jianbo Liu Reviewed-by: Oz Shlomo Reviewed-by: Roi Dayan Signed-off-by: Saeed Mahameed Signed-off-by: Greg Kroah-Hartman --- .../mellanox/mlx5/core/eswitch_offloads_termtbl.c | 22 ++++++++++++++++------ 1 file changed, 16 insertions(+), 6 deletions(-) (limited to 'drivers/net/ethernet/mellanox') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c index 1d55a324a17e..7879e1746297 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c @@ -177,22 +177,32 @@ mlx5_eswitch_termtbl_actions_move(struct mlx5_flow_act *src, memset(&src->vlan[1], 0, sizeof(src->vlan[1])); } +static bool mlx5_eswitch_offload_is_uplink_port(const struct mlx5_eswitch *esw, + const struct mlx5_flow_spec *spec) +{ + u32 port_mask, port_value; + + if (MLX5_CAP_ESW_FLOWTABLE(esw->dev, flow_source)) + return spec->flow_context.flow_source == MLX5_VPORT_UPLINK; + + port_mask = MLX5_GET(fte_match_param, spec->match_criteria, + misc_parameters.source_port); + port_value = MLX5_GET(fte_match_param, spec->match_value, + misc_parameters.source_port); + return (port_mask & port_value & 0xffff) == MLX5_VPORT_UPLINK; +} + bool mlx5_eswitch_termtbl_required(struct mlx5_eswitch *esw, struct mlx5_flow_act *flow_act, struct mlx5_flow_spec *spec) { - u32 port_mask = MLX5_GET(fte_match_param, spec->match_criteria, - misc_parameters.source_port); - u32 port_value = MLX5_GET(fte_match_param, spec->match_value, - misc_parameters.source_port); - if (!MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, termination_table)) return false; /* push vlan on RX */ return (flow_act->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH) && - ((port_mask & port_value) == MLX5_VPORT_UPLINK); + mlx5_eswitch_offload_is_uplink_port(esw, spec); } struct mlx5_flow_handle * -- cgit v1.2.3 From ee50c8f5dbb4d1068d1d28b10355da3e0e4036e4 Mon Sep 17 00:00:00 2001 From: Dmytro Linkin Date: Thu, 29 Aug 2019 15:24:27 +0000 Subject: net/mlx5e: Remove incorrect match criteria assignment line [ Upstream commit 752d3dc06d6936d5a357a18b6b51d91c7e134e88 ] Driver have function, which enable match criteria for misc parameters in dependence of eswitch capabilities. Fixes: 4f5d1beadc10 ("Merge branch 'mlx5-next' of git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux") Signed-off-by: Dmytro Linkin Reviewed-by: Jianbo Liu Reviewed-by: Roi Dayan Reviewed-by: Saeed Mahameed Signed-off-by: Saeed Mahameed Signed-off-by: Greg Kroah-Hartman --- drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 1 - 1 file changed, 1 deletion(-) (limited to 'drivers/net/ethernet/mellanox') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 0323fd078271..35945cdd0a61 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -285,7 +285,6 @@ mlx5_eswitch_add_fwd_rule(struct mlx5_eswitch *esw, mlx5_eswitch_set_rule_source_port(esw, spec, attr); - spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS; if (attr->outer_match_level != MLX5_MATCH_NONE) spec->match_criteria_enable |= MLX5_MATCH_OUTER_HEADERS; -- cgit v1.2.3 From 3001dbfa21bc4f365bf6063a8df30888ffff4dcf Mon Sep 17 00:00:00 2001 From: Aya Levin Date: Wed, 23 Oct 2019 12:57:54 +0300 Subject: net/mlx5e: Initialize on stack link modes bitmap [ Upstream commit 926b37f76fb0a22fe93c8873c819fd167180e85c ] Initialize link modes bitmap on stack before using it, otherwise the outcome of ethtool set link ksettings might have unexpected values. Fixes: 4b95840a6ced ("net/mlx5e: Fix matching of speed to PRM link modes") Signed-off-by: Aya Levin Signed-off-by: Saeed Mahameed Signed-off-by: Greg Kroah-Hartman --- drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/ethernet/mellanox') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 20e628c907e5..a9bb8e2b34a7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -1021,7 +1021,7 @@ static bool ext_link_mode_requested(const unsigned long *adver) { #define MLX5E_MIN_PTYS_EXT_LINK_MODE_BIT ETHTOOL_LINK_MODE_50000baseKR_Full_BIT int size = __ETHTOOL_LINK_MODE_MASK_NBITS - MLX5E_MIN_PTYS_EXT_LINK_MODE_BIT; - __ETHTOOL_DECLARE_LINK_MODE_MASK(modes); + __ETHTOOL_DECLARE_LINK_MODE_MASK(modes) = {0,}; bitmap_set(modes, MLX5E_MIN_PTYS_EXT_LINK_MODE_BIT, size); return bitmap_intersects(modes, adver, __ETHTOOL_LINK_MODE_MASK_NBITS); -- cgit v1.2.3 From 551486fb7909e4f4507e750d3ec2ca7a05919f62 Mon Sep 17 00:00:00 2001 From: Parav Pandit Date: Thu, 19 Sep 2019 15:58:14 -0500 Subject: net/mlx5: Fix rtable reference leak [ Upstream commit 2347cee83b2bd868bde2d283db0fac89f22be4e0 ] If the rt entry gateway family is not AF_INET for multipath device, rtable reference is leaked. Hence, fix it by releasing the reference. Fixes: 5fb091e8130b ("net/mlx5e: Use hint to resolve route when in HW multipath mode") Fixes: e32ee6c78efa ("net/mlx5e: Support tunnel encap over tagged Ethernet") Signed-off-by: Parav Pandit Signed-off-by: Saeed Mahameed Signed-off-by: Greg Kroah-Hartman --- drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) (limited to 'drivers/net/ethernet/mellanox') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c index a6a52806be45..310f65ef5446 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c @@ -90,15 +90,19 @@ static int mlx5e_route_lookup_ipv4(struct mlx5e_priv *priv, if (ret) return ret; - if (mlx5_lag_is_multipath(mdev) && rt->rt_gw_family != AF_INET) + if (mlx5_lag_is_multipath(mdev) && rt->rt_gw_family != AF_INET) { + ip_rt_put(rt); return -ENETUNREACH; + } #else return -EOPNOTSUPP; #endif ret = get_route_and_out_devs(priv, rt->dst.dev, route_dev, out_dev); - if (ret < 0) + if (ret < 0) { + ip_rt_put(rt); return ret; + } if (!(*out_ttl)) *out_ttl = ip4_dst_hoplimit(&rt->dst); @@ -142,8 +146,10 @@ static int mlx5e_route_lookup_ipv6(struct mlx5e_priv *priv, *out_ttl = ip6_dst_hoplimit(dst); ret = get_route_and_out_devs(priv, dst->dev, route_dev, out_dev); - if (ret < 0) + if (ret < 0) { + dst_release(dst); return ret; + } #else return -EOPNOTSUPP; #endif -- cgit v1.2.3 From 0e22b9051704c12e9daffc01d40389b044549226 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Wed, 30 Oct 2019 11:04:22 +0200 Subject: mlxsw: core: Unpublish devlink parameters during reload [ Upstream commit b7265a0df82c1716bf788096217083ed65a8bb14 ] The devlink parameter "acl_region_rehash_interval" is a runtime parameter whose value is stored in a dynamically allocated memory. While reloading the driver, this memory is freed and then allocated again. A use-after-free might happen if during this time frame someone tries to retrieve its value. Since commit 070c63f20f6c ("net: devlink: allow to change namespaces during reload") the use-after-free can be reliably triggered when reloading the driver into a namespace, as after freeing the memory (via reload_down() callback) all the parameters are notified. Fix this by unpublishing and then re-publishing the parameters during reload. Fixes: 98bbf70c1c41 ("mlxsw: spectrum: add "acl_region_rehash_interval" devlink param") Fixes: 7c62cfb8c574 ("devlink: publish params only after driver init is done") Signed-off-by: Jiri Pirko Signed-off-by: Ido Schimmel Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- drivers/net/ethernet/mellanox/mlxsw/core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/net/ethernet/mellanox') diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c index 17ceac7505e5..b94cdbd7bb18 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.c +++ b/drivers/net/ethernet/mellanox/mlxsw/core.c @@ -1128,7 +1128,7 @@ __mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info, if (err) goto err_thermal_init; - if (mlxsw_driver->params_register && !reload) + if (mlxsw_driver->params_register) devlink_params_publish(devlink); return 0; @@ -1201,7 +1201,7 @@ void mlxsw_core_bus_device_unregister(struct mlxsw_core *mlxsw_core, return; } - if (mlxsw_core->driver->params_unregister && !reload) + if (mlxsw_core->driver->params_unregister) devlink_params_unpublish(devlink); mlxsw_thermal_fini(mlxsw_core->thermal); mlxsw_hwmon_fini(mlxsw_core->hwmon); -- cgit v1.2.3 From 8135afc2cecc434c0fc3eee0d85446def2389d71 Mon Sep 17 00:00:00 2001 From: Aya Levin Date: Wed, 2 Oct 2019 16:53:21 +0300 Subject: net/mlx5e: Fix ethtool self test: link speed [ Upstream commit 534e7366f41b0c689b01af4375aefcd1462adedf ] Ethtool self test contains a test for link speed. This test reads the PTYS register and determines whether the current speed is valid or not. Change current implementation to use the function mlx5e_port_linkspeed() that does the same check and fails when speed is invalid. This code redundancy lead to a bug when mlx5e_port_linkspeed() was updated with expended speeds and the self test was not. Fixes: 2c81bfd5ae56 ("net/mlx5e: Move port speed code from en_ethtool.c to en/port.c") Signed-off-by: Aya Levin Reviewed-by: Moshe Shemesh Signed-off-by: Saeed Mahameed Signed-off-by: Greg Kroah-Hartman --- drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c | 15 +++------------ 1 file changed, 3 insertions(+), 12 deletions(-) (limited to 'drivers/net/ethernet/mellanox') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c b/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c index 840ec945ccba..bbff8d8ded76 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c @@ -35,6 +35,7 @@ #include #include #include "en.h" +#include "en/port.h" enum { MLX5E_ST_LINK_STATE, @@ -80,22 +81,12 @@ static int mlx5e_test_link_state(struct mlx5e_priv *priv) static int mlx5e_test_link_speed(struct mlx5e_priv *priv) { - u32 out[MLX5_ST_SZ_DW(ptys_reg)]; - u32 eth_proto_oper; - int i; + u32 speed; if (!netif_carrier_ok(priv->netdev)) return 1; - if (mlx5_query_port_ptys(priv->mdev, out, sizeof(out), MLX5_PTYS_EN, 1)) - return 1; - - eth_proto_oper = MLX5_GET(ptys_reg, out, eth_proto_oper); - for (i = 0; i < MLX5E_LINK_MODES_NUMBER; i++) { - if (eth_proto_oper & MLX5E_PROT_MASK(i)) - return 0; - } - return 1; + return mlx5e_port_linkspeed(priv->mdev, &speed); } struct mlx5ehdr { -- cgit v1.2.3 From 372a4ffae4c3ea164a0ef57ca02b7e75b5b24011 Mon Sep 17 00:00:00 2001 From: Maxim Mikityanskiy Date: Mon, 16 Sep 2019 14:54:20 +0300 Subject: net/mlx5e: Fix handling of compressed CQEs in case of low NAPI budget [ Upstream commit 9df86bdb6746d7fcfc2fda715f7a7c3d0ddb2654 ] When CQE compression is enabled, compressed CQEs use the following structure: a title is followed by one or many blocks, each containing 8 mini CQEs (except the last, which may contain fewer mini CQEs). Due to NAPI budget restriction, a complete structure is not always parsed in one NAPI run, and some blocks with mini CQEs may be deferred to the next NAPI poll call - we have the mlx5e_decompress_cqes_cont call in the beginning of mlx5e_poll_rx_cq. However, if the budget is extremely low, some blocks may be left even after that, but the code that follows the mlx5e_decompress_cqes_cont call doesn't check it and assumes that a new CQE begins, which may not be the case. In such cases, random memory corruptions occur. An extremely low NAPI budget of 8 is used when busy_poll or busy_read is active. This commit adds a check to make sure that the previous compressed CQE has been completely parsed after mlx5e_decompress_cqes_cont, otherwise it prevents a new CQE from being fetched in the middle of a compressed CQE. This commit fixes random crashes in __build_skb, __page_pool_put_page and other not-related-directly places, that used to happen when both CQE compression and busy_poll/busy_read were enabled. Fixes: 7219ab34f184 ("net/mlx5e: CQE compression") Signed-off-by: Maxim Mikityanskiy Signed-off-by: Saeed Mahameed Signed-off-by: Greg Kroah-Hartman --- drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'drivers/net/ethernet/mellanox') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index ac6e586d403d..fb139f8b9acf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -1367,8 +1367,11 @@ int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget) if (unlikely(!test_bit(MLX5E_RQ_STATE_ENABLED, &rq->state))) return 0; - if (rq->cqd.left) + if (rq->cqd.left) { work_done += mlx5e_decompress_cqes_cont(rq, cqwq, 0, budget); + if (rq->cqd.left || work_done >= budget) + goto out; + } cqe = mlx5_cqwq_get_cqe(cqwq); if (!cqe) { -- cgit v1.2.3