From 59079438a664559bb1f6f5fe85e306962ef9286e Mon Sep 17 00:00:00 2001 From: Mikhael Goikhman Date: Wed, 10 Mar 2021 23:09:09 -0800 Subject: net/mlx5: Remove unused mlx5_core_health member recover_work The code related to health->recover_work was removed in commit 63cbc552eebf ("net/mlx5: Handle SW reset of FW in error flow") Fix struct mlx5_core_health accordingly. Signed-off-by: Mikhael Goikhman Reviewed-by: Tariq Toukan Signed-off-by: Saeed Mahameed --- include/linux/mlx5/driver.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 53b89631a1d9..8fe51b4a781e 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -438,7 +438,6 @@ struct mlx5_core_health { unsigned long flags; struct work_struct fatal_report_work; struct work_struct report_work; - struct delayed_work recover_work; struct devlink_health_reporter *fw_reporter; struct devlink_health_reporter *fw_fatal_reporter; }; -- cgit From 59c904c8fffd903c1dae5fc6a402b88fa6dfc874 Mon Sep 17 00:00:00 2001 From: Mark Bloch Date: Wed, 10 Mar 2021 23:09:11 -0800 Subject: net/mlx5: E-Switch, Add eswitch pointer to each representor Store the managing E-Switch of each representor. This will be used when a representor is created on eswitch manager 0 but the vport belongs to eswitch manager 1. Signed-off-by: Mark Bloch Reviewed-by: Saeed Mahameed Signed-off-by: Saeed Mahameed --- include/linux/mlx5/eswitch.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/mlx5/eswitch.h b/include/linux/mlx5/eswitch.h index 994c2c8cb4fd..72d480df2a03 100644 --- a/include/linux/mlx5/eswitch.h +++ b/include/linux/mlx5/eswitch.h @@ -48,6 +48,7 @@ struct mlx5_eswitch_rep { /* Only IB rep is using vport_index */ u16 vport_index; u32 vlan_refcount; + struct mlx5_eswitch *esw; }; void mlx5_eswitch_register_vport_reps(struct mlx5_eswitch *esw, -- cgit From 3a46f4fb55ffd46e475e3fc53b1252f722cf647e Mon Sep 17 00:00:00 2001 From: Mark Bloch Date: Wed, 10 Mar 2021 23:09:13 -0800 Subject: net/mlx5: E-Switch, Refactor send to vport to be more generic Now that each representor stores a pointer to the managing E-Switch use that information when creating the send-to-vport rules. Signed-off-by: Mark Bloch Reviewed-by: Saeed Mahameed Signed-off-by: Saeed Mahameed --- include/linux/mlx5/eswitch.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mlx5/eswitch.h b/include/linux/mlx5/eswitch.h index 72d480df2a03..2ec0527991c8 100644 --- a/include/linux/mlx5/eswitch.h +++ b/include/linux/mlx5/eswitch.h @@ -62,8 +62,8 @@ struct mlx5_eswitch_rep *mlx5_eswitch_vport_rep(struct mlx5_eswitch *esw, u16 vport_num); void *mlx5_eswitch_uplink_get_proto_dev(struct mlx5_eswitch *esw, u8 rep_type); struct mlx5_flow_handle * -mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, - u16 vport_num, u32 sqn); +mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *on_esw, + struct mlx5_eswitch_rep *rep, u32 sqn); u16 mlx5_eswitch_get_total_vports(const struct mlx5_core_dev *dev); -- cgit From c3e666f1ada9cbfbe5465f122f9a2d63ddfd25ed Mon Sep 17 00:00:00 2001 From: Mark Bloch Date: Wed, 10 Mar 2021 23:09:14 -0800 Subject: net/mlx5: Add IFC bits needed for single FDB mode Currently we operate in a mode where each eswitch manager has a separate FDB. In order to combine these multiple FDBs we expose new caps to allow this: - Set root flow table which isn't native. - Set FDB a different selection mode when in LAG mode. Signed-off-by: Mark Bloch Reviewed-by: Saeed Mahameed Signed-off-by: Saeed Mahameed --- include/linux/mlx5/mlx5_ifc.h | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index df5d91c8b2d4..3ee7a86f39e4 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -806,9 +806,11 @@ struct mlx5_ifc_e_switch_cap_bits { u8 vport_svlan_insert[0x1]; u8 vport_cvlan_insert_if_not_exist[0x1]; u8 vport_cvlan_insert_overwrite[0x1]; - u8 reserved_at_5[0x3]; + u8 reserved_at_5[0x2]; + u8 esw_shared_ingress_acl[0x1]; u8 esw_uplink_ingress_acl[0x1]; - u8 reserved_at_9[0x10]; + u8 root_ft_on_other_esw[0x1]; + u8 reserved_at_a[0xf]; u8 esw_functions_changed[0x1]; u8 reserved_at_1a[0x1]; u8 ecpf_vport_exists[0x1]; @@ -1502,7 +1504,8 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 reserved_at_270[0x6]; u8 lag_dct[0x2]; u8 lag_tx_port_affinity[0x1]; - u8 reserved_at_279[0x2]; + u8 lag_native_fdb_selection[0x1]; + u8 reserved_at_27a[0x1]; u8 lag_master[0x1]; u8 num_lag_ports[0x4]; @@ -10036,14 +10039,19 @@ struct mlx5_ifc_set_flow_table_root_in_bits { u8 reserved_at_60[0x20]; u8 table_type[0x8]; - u8 reserved_at_88[0x18]; + u8 reserved_at_88[0x7]; + u8 table_of_other_vport[0x1]; + u8 table_vport_number[0x10]; u8 reserved_at_a0[0x8]; u8 table_id[0x18]; u8 reserved_at_c0[0x8]; u8 underlay_qpn[0x18]; - u8 reserved_at_e0[0x120]; + u8 table_eswitch_owner_vhca_id_valid[0x1]; + u8 reserved_at_e1[0xf]; + u8 table_eswitch_owner_vhca_id[0x10]; + u8 reserved_at_100[0x100]; }; enum { @@ -10273,7 +10281,8 @@ struct mlx5_ifc_dcbx_param_bits { }; struct mlx5_ifc_lagc_bits { - u8 reserved_at_0[0x1d]; + u8 fdb_selection_mode[0x1]; + u8 reserved_at_1[0x1c]; u8 lag_state[0x3]; u8 reserved_at_20[0x14]; -- cgit From 26bf30902c10473ba38f220d3401a61c56d8db3b Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Wed, 10 Mar 2021 23:09:15 -0800 Subject: net/mlx5: Use order-0 allocations for EQs Currently we are allocating high-order page for EQs. In case of fragmented system, VF hot remove/add in VMs for example, there isn't enough contiguous memory for EQs allocation, which results in crashing of the VM. Therefore, use order-0 fragments for the EQ allocations instead. Performance tests: ConnectX-5 100Gbps, CPU: Intel(R) Xeon(R) CPU E5-2697 v3 @ 2.60GHz Performance tests show no sensible degradation. Signed-off-by: Tariq Toukan Signed-off-by: Shay Drory Signed-off-by: Saeed Mahameed --- include/linux/mlx5/driver.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/linux') diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 8fe51b4a781e..5c0422930b01 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -873,6 +873,11 @@ static inline u32 mlx5_base_mkey(const u32 key) return key & 0xffffff00u; } +static inline u32 wq_get_byte_sz(u8 log_sz, u8 log_stride) +{ + return ((u32)1 << log_sz) << log_stride; +} + static inline void mlx5_init_fbc_offset(struct mlx5_buf_list *frags, u8 log_stride, u8 log_sz, u16 strides_offset, -- cgit From c3d5c2d96d69f2578d6fbf66e39cc2cf840d9812 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Sun, 4 Apr 2021 10:22:18 +0300 Subject: PCI/IOV: Add sysfs MSI-X vector assignment interface A typical cloud provider SR-IOV use case is to create many VFs for use by guest VMs. The VFs may not be assigned to a VM until a customer requests a VM of a certain size, e.g., number of CPUs. A VF may need MSI-X vectors proportional to the number of CPUs in the VM, but there is no standard way to change the number of MSI-X vectors supported by a VF. Some Mellanox ConnectX devices support dynamic assignment of MSI-X vectors to SR-IOV VFs. This can be done by the PF driver after VFs are enabled, and it can be done without affecting VFs that are already in use. The hardware supports a limited pool of MSI-X vectors that can be assigned to the PF or to individual VFs. This is device-specific behavior that requires support in the PF driver. Add a read-only "sriov_vf_total_msix" sysfs file for the PF and a writable "sriov_vf_msix_count" file for each VF. Management software may use these to learn how many MSI-X vectors are available and to dynamically assign them to VFs before the VFs are passed through to a VM. If the PF driver implements the ->sriov_get_vf_total_msix() callback, "sriov_vf_total_msix" contains the total number of MSI-X vectors available for distribution among VFs. If no driver is bound to the VF, writing "N" to "sriov_vf_msix_count" uses the PF driver ->sriov_set_msix_vec_count() callback to assign "N" MSI-X vectors to the VF. When a VF driver subsequently reads the MSI-X Message Control register, it will see the new Table Size "N". Link: https://lore.kernel.org/linux-pci/20210314124256.70253-2-leon@kernel.org Acked-by: Bjorn Helgaas Signed-off-by: Leon Romanovsky --- include/linux/pci.h | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pci.h b/include/linux/pci.h index 86c799c97b77..9b575a676888 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -856,6 +856,12 @@ struct module; * e.g. drivers/net/e100.c. * @sriov_configure: Optional driver callback to allow configuration of * number of VFs to enable via sysfs "sriov_numvfs" file. + * @sriov_set_msix_vec_count: PF Driver callback to change number of MSI-X + * vectors on a VF. Triggered via sysfs "sriov_vf_msix_count". + * This will change MSI-X Table Size in the VF Message Control + * registers. + * @sriov_get_vf_total_msix: PF driver callback to get the total number of + * MSI-X vectors available for distribution to the VFs. * @err_handler: See Documentation/PCI/pci-error-recovery.rst * @groups: Sysfs attribute groups. * @driver: Driver model structure. @@ -871,6 +877,8 @@ struct pci_driver { int (*resume)(struct pci_dev *dev); /* Device woken up */ void (*shutdown)(struct pci_dev *dev); int (*sriov_configure)(struct pci_dev *dev, int num_vfs); /* On PF */ + int (*sriov_set_msix_vec_count)(struct pci_dev *vf, int msix_vec_count); /* On PF */ + u32 (*sriov_get_vf_total_msix)(struct pci_dev *pf); const struct pci_error_handlers *err_handler; const struct attribute_group **groups; struct device_driver driver; -- cgit From 0b989c1e37053196676b2238f82195bd5a339d58 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Sun, 14 Mar 2021 14:42:54 +0200 Subject: net/mlx5: Add dynamic MSI-X capabilities bits These new fields declare the number of MSI-X vectors that is possible to allocate on the VF through PF configuration. Value must be in range defined by min_dynamic_vf_msix_table_size and max_dynamic_vf_msix_table_size. The driver should continue to query its MSI-X table through PCI configuration header. Link: https://lore.kernel.org/linux-pci/20210314124256.70253-3-leon@kernel.org Acked-by: Bjorn Helgaas Signed-off-by: Leon Romanovsky --- include/linux/mlx5/mlx5_ifc.h | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 3ee7a86f39e4..432290b58a0b 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -1683,7 +1683,16 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 reserved_at_6e0[0x10]; u8 sf_base_id[0x10]; - u8 reserved_at_700[0x80]; + u8 reserved_at_700[0x8]; + u8 num_total_dynamic_vf_msix[0x18]; + u8 reserved_at_720[0x14]; + u8 dynamic_msix_table_size[0xc]; + u8 reserved_at_740[0xc]; + u8 min_dynamic_vf_msix_table_size[0x4]; + u8 reserved_at_750[0x4]; + u8 max_dynamic_vf_msix_table_size[0xc]; + + u8 reserved_at_760[0x20]; u8 vhca_tunnel_commands[0x40]; u8 reserved_at_7c0[0x40]; }; -- cgit