Add virtual function device ids. Setting VF flag to device private data Establish comm channel when sriov is enabled, and boot vfs through it. Each slave gets one msi-X for completions, the master also gets one for async events.
Signed-off-by: Liran Liss <[email protected]> Signed-off-by: Yevgeny Petrilin <[email protected]> --- drivers/net/mlx4/eq.c | 5 +- drivers/net/mlx4/main.c | 307 +++++++++++++++++++++++++++++++++++----------- drivers/net/mlx4/mlx4.h | 4 + 3 files changed, 240 insertions(+), 76 deletions(-) diff --git a/drivers/net/mlx4/eq.c b/drivers/net/mlx4/eq.c index 1cb692d..9126c8e 100644 --- a/drivers/net/mlx4/eq.c +++ b/drivers/net/mlx4/eq.c @@ -805,13 +805,14 @@ int mlx4_init_eq_table(struct mlx4_dev *dev) priv->eq_table.eq[dev->caps.num_comp_vectors].eqn, err); } - for (i = 0; i < dev->caps.num_comp_vectors + 1; ++i) + for (i = 0; i < dev->caps.num_comp_vectors + !(mlx4_is_slave(dev)); ++i) eq_set_ci(&priv->eq_table.eq[i], 1); return 0; err_out_async: - mlx4_free_eq(dev, &priv->eq_table.eq[dev->caps.num_comp_vectors]); + if (!mlx4_is_slave(dev)) + mlx4_free_eq(dev, &priv->eq_table.eq[dev->caps.num_comp_vectors]); err_out_comp: i = dev->caps.num_comp_vectors; diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c index f67f992..3331c33 100644 --- a/drivers/net/mlx4/main.c +++ b/drivers/net/mlx4/main.c @@ -74,6 +74,23 @@ MODULE_PARM_DESC(msi_x, "attempt to use MSI-X if nonzero"); #endif /* CONFIG_PCI_MSI */ +#ifdef CONFIG_PCI_IOV + +static int sr_iov; +module_param(sr_iov, int, 0444); +MODULE_PARM_DESC(sr_iov, "enable #sr_iov functions if sr_iov > 0"); + +static int probe_vf; +module_param(probe_vf, int, 0444); +MODULE_PARM_DESC(probe_vf, "number of vfs to probe by pf driver (sr_iov > 0)"); + +#else /* CONFIG_PCI_IOV */ + +#define sr_iov 0 +#define probe_vf 0 + +#endif /* CONFIG_PCI_IOV */ + static char mlx4_version[] __devinitdata = DRV_NAME ": Mellanox ConnectX core driver v" DRV_VERSION " (" DRV_RELDATE ")\n"; @@ -780,12 +797,56 @@ static void mlx4_free_icms(struct mlx4_dev *dev) mlx4_free_icm(dev, priv->fw.aux_icm, 0); } +static void mlx4_slave_exit(struct mlx4_dev *dev) +{ + struct mlx4_priv *priv = mlx4_priv(dev); + + down(&priv->cmd.poll_sem); + if (mlx4_comm_cmd(dev, MLX4_COMM_CMD_RESET, 0, MLX4_COMM_TIME)) + mlx4_warn(dev, "Failed to close slave function.\n"); + up(&priv->cmd.poll_sem); +} + static void mlx4_close_hca(struct mlx4_dev *dev) { - mlx4_CLOSE_HCA(dev, 0); - mlx4_free_icms(dev); - mlx4_UNMAP_FA(dev); - mlx4_free_icm(dev, mlx4_priv(dev)->fw.fw_icm, 0); + if (mlx4_is_slave(dev)) + mlx4_slave_exit(dev); + else { + mlx4_CLOSE_HCA(dev, 0); + mlx4_free_icms(dev); + mlx4_UNMAP_FA(dev); + mlx4_free_icm(dev, mlx4_priv(dev)->fw.fw_icm, 0); + } +} + +static int mlx4_init_slave(struct mlx4_dev *dev) +{ + struct mlx4_priv *priv = mlx4_priv(dev); + u64 dma = (u64) priv->mfunc.vhcr_dma; + + down(&priv->cmd.poll_sem); + mlx4_warn(dev, "Sending reset\n"); + if (mlx4_comm_cmd(dev, MLX4_COMM_CMD_RESET, 0, MLX4_COMM_TIME)) + goto err; + mlx4_warn(dev, "Sending vhcr0\n"); + if (mlx4_comm_cmd(dev, MLX4_COMM_CMD_VHCR0, dma >> 48, + MLX4_COMM_TIME)) + goto err; + if (mlx4_comm_cmd(dev, MLX4_COMM_CMD_VHCR1, dma >> 32, + MLX4_COMM_TIME)) + goto err; + if (mlx4_comm_cmd(dev, MLX4_COMM_CMD_VHCR2, dma >> 16, + MLX4_COMM_TIME)) + goto err; + if (mlx4_comm_cmd(dev, MLX4_COMM_CMD_VHCR_EN, dma, MLX4_COMM_TIME)) + goto err; + up(&priv->cmd.poll_sem); + return 0; + +err: + mlx4_comm_cmd(dev, MLX4_COMM_CMD_RESET, 0, 0); + up(&priv->cmd.poll_sem); + return -EIO; } static int mlx4_init_hca(struct mlx4_dev *dev) @@ -799,51 +860,65 @@ static int mlx4_init_hca(struct mlx4_dev *dev) u64 icm_size; int err; - err = mlx4_QUERY_FW(dev); - if (err) { - if (err == -EACCES) - mlx4_info(dev, "non-primary physical function, skipping.\n"); - else - mlx4_err(dev, "QUERY_FW command failed, aborting.\n"); - return err; - } + if (!mlx4_is_slave(dev)) { + err = mlx4_QUERY_FW(dev); + if (err) { + if (err == -EACCES) + mlx4_info(dev, "non-primary physical function, skipping.\n"); + else + mlx4_err(dev, "QUERY_FW command failed, aborting.\n"); + return err; + } - err = mlx4_load_fw(dev); - if (err) { - mlx4_err(dev, "Failed to start FW, aborting.\n"); - return err; - } + err = mlx4_load_fw(dev); + if (err) { + mlx4_err(dev, "Failed to start FW, aborting.\n"); + return err; + } - mlx4_cfg.log_pg_sz_m = 1; - mlx4_cfg.log_pg_sz = 0; - err = mlx4_MOD_STAT_CFG(dev, &mlx4_cfg); - if (err) - mlx4_warn(dev, "Failed to override log_pg_sz parameter\n"); + mlx4_cfg.log_pg_sz_m = 1; + mlx4_cfg.log_pg_sz = 0; + err = mlx4_MOD_STAT_CFG(dev, &mlx4_cfg); + if (err) + mlx4_warn(dev, "Failed to override log_pg_sz parameter\n"); - err = mlx4_dev_cap(dev, &dev_cap); - if (err) { - mlx4_err(dev, "QUERY_DEV_CAP command failed, aborting.\n"); - goto err_stop_fw; - } + err = mlx4_dev_cap(dev, &dev_cap); + if (err) { + mlx4_err(dev, "QUERY_DEV_CAP command failed, aborting.\n"); + goto err_stop_fw; + } - profile = default_profile; + profile = default_profile; - icm_size = mlx4_make_profile(dev, &profile, &dev_cap, &init_hca); - if ((long long) icm_size < 0) { - err = icm_size; - goto err_stop_fw; - } + icm_size = mlx4_make_profile(dev, &profile, &dev_cap, &init_hca); + if ((long long) icm_size < 0) { + err = icm_size; + goto err_stop_fw; + } - init_hca.log_uar_sz = ilog2(dev->caps.num_uars); + init_hca.log_uar_sz = ilog2(dev->caps.num_uars); - err = mlx4_init_icm(dev, &dev_cap, &init_hca, icm_size); - if (err) - goto err_stop_fw; + err = mlx4_init_icm(dev, &dev_cap, &init_hca, icm_size); + if (err) + goto err_stop_fw; - err = mlx4_INIT_HCA(dev, &init_hca); - if (err) { - mlx4_err(dev, "INIT_HCA command failed, aborting.\n"); - goto err_free_icm; + err = mlx4_INIT_HCA(dev, &init_hca); + if (err) { + mlx4_err(dev, "INIT_HCA command failed, aborting.\n"); + goto err_free_icm; + } + } else { + err = mlx4_init_slave(dev); + if (err) { + mlx4_err(dev, "Failed to initialize slave\n"); + return err; + } + + err = mlx4_slave_cap(dev); + if (err) { + mlx4_err(dev, "Failed to obtain slave caps\n"); + goto err_close; + } } err = mlx4_QUERY_ADAPTER(dev, &adapter); @@ -858,15 +933,17 @@ static int mlx4_init_hca(struct mlx4_dev *dev) return 0; err_close: - mlx4_CLOSE_HCA(dev, 0); + mlx4_close_hca(dev); err_free_icm: - mlx4_free_icms(dev); + if (!mlx4_is_slave(dev)) + mlx4_free_icms(dev); err_stop_fw: - mlx4_UNMAP_FA(dev); - mlx4_free_icm(dev, priv->fw.fw_icm, 0); - + if (!mlx4_is_slave(dev)) { + mlx4_UNMAP_FA(dev); + mlx4_free_icm(dev, priv->fw.fw_icm, 0); + } return err; } @@ -1041,8 +1118,13 @@ static void mlx4_enable_msi_x(struct mlx4_dev *dev) int i; if (msi_x) { - nreq = min_t(int, dev->caps.num_eqs - dev->caps.reserved_eqs, - num_possible_cpus() + 1); + /* The master only uses en event EQ, + * Each one of the slaves have 1 completion eq */ + if (mlx4_is_mfunc(dev)) + nreq = 1 + !!mlx4_is_master(dev); + else + nreq = min_t(int, dev->caps.num_eqs - dev->caps.reserved_eqs, + num_possible_cpus() + 1); entries = kcalloc(nreq, sizeof *entries, GFP_KERNEL); if (!entries) goto no_msi; @@ -1137,10 +1219,10 @@ static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id) } /* - * Check for BARs. We expect 0: 1MB + * Check for BARs. */ - if (!(pci_resource_flags(pdev, 0) & IORESOURCE_MEM) || - pci_resource_len(pdev, 0) != 1 << 20) { + if (((id == NULL) || !(id->driver_data & MLX4_VF)) && + !(pci_resource_flags(pdev, 0) & IORESOURCE_MEM)) { dev_err(&pdev->dev, "Missing DCS, aborting.\n"); err = -ENODEV; goto err_disable_pdev; @@ -1198,34 +1280,83 @@ static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id) INIT_LIST_HEAD(&priv->pgdir_list); mutex_init(&priv->pgdir_mutex); - /* - * Now reset the HCA before we touch the PCI capabilities or - * attempt a firmware command, since a boot ROM may have left - * the HCA in an undefined state. - */ - err = mlx4_reset(dev); - if (err) { - mlx4_err(dev, "Failed to reset HCA, aborting.\n"); - goto err_free_dev; + /* Detect if this device is a virtual function */ + if (id && id->driver_data & MLX4_VF) { + /* When acting as pf, we normally skip vfs unless explicitly + * requested to probe them. */ + if (sr_iov && PCI_FUNC(pdev->devfn) > probe_vf) { + mlx4_warn(dev, "Skipping virtual function:%d\n", + PCI_FUNC(pdev->devfn)); + err = -ENODEV; + goto err_free_dev; + } + mlx4_warn(dev, "Detected virtual function - running in slave mode\n"); + dev->flags |= MLX4_FLAG_SLAVE; + } + + /* We reset the device and enable SRIOV only for physical devices */ + if (!mlx4_is_slave(dev)) { + /* + * Now reset the HCA before we touch the PCI capabilities or + * attempt a firmware command, since a boot ROM may have left + * the HCA in an undefined state. + */ + err = mlx4_reset(dev); + if (err) { + mlx4_err(dev, "Failed to reset HCA, aborting.\n"); + goto err_free_dev; + } + if (sr_iov) { + mlx4_warn(dev, "Enabling sriov with:%d vfs\n", sr_iov); + if (pci_enable_sriov(pdev, sr_iov)) { + mlx4_err(dev, "Failed to enable sriov, aborting.\n"); + goto err_free_dev; + } + mlx4_warn(dev, "Running in master mode\n"); + dev->flags |= MLX4_FLAG_SRIOV | MLX4_FLAG_MASTER; + } } if (mlx4_cmd_init(dev)) { mlx4_err(dev, "Failed to init command interface, aborting.\n"); - goto err_free_dev; + goto err_sriov; + } + + /* In slave functions, the communication channel must be initialized before + * posting commands */ + if (mlx4_is_slave(dev)) { + if (mlx4_multi_func_init(dev)) { + mlx4_err(dev, "Failed to init slave mfunc interface, aborting.\n"); + goto err_cmd; + } } err = mlx4_init_hca(dev); if (err) goto err_cmd; + /* In master functions, the communication channel must be initialized after obtaining + * its address from fw */ + if (mlx4_is_master(dev)) { + dev->num_slaves = MLX4_MAX_NUM_SLAVES; + if (mlx4_multi_func_init(dev)) { + mlx4_err(dev, "Failed to init master mfunc interface, aborting.\n"); + goto err_close; + } + } + err = mlx4_alloc_eq_table(dev); if (err) goto err_close; mlx4_enable_msi_x(dev); + if (mlx4_is_slave(dev) && !(dev->flags & MLX4_FLAG_MSI_X)) { + mlx4_err(dev, "INTx is not supported in slave mode, aborting.\n"); + goto err_free_eq; + } err = mlx4_setup_hca(dev); - if (err == -EBUSY && (dev->flags & MLX4_FLAG_MSI_X)) { + if (err == -EBUSY && (dev->flags & MLX4_FLAG_MSI_X) && !mlx4_is_slave(dev)) { dev->flags &= ~MLX4_FLAG_MSI_X; pci_disable_msix(pdev); err = mlx4_setup_hca(dev); @@ -1284,6 +1415,12 @@ err_close: err_cmd: mlx4_cmd_cleanup(dev); +err_sriov: + if (mlx4_is_mfunc(dev)) + mlx4_multi_func_cleanup(dev); + if (sr_iov && (dev->flags & MLX4_FLAG_SRIOV)) + pci_disable_sriov(pdev); + err_free_dev: kfree(priv); @@ -1316,6 +1453,9 @@ static void mlx4_remove_one(struct pci_dev *pdev) int p; if (dev) { + /* Stop serving commands and events over comm channel */ + if (mlx4_is_mfunc(dev)) + cancel_delayed_work_sync(&priv->mfunc.comm_work); mlx4_stop_sense(dev); mlx4_unregister_device(dev); @@ -1339,10 +1479,16 @@ static void mlx4_remove_one(struct pci_dev *pdev) mlx4_cleanup_uar_table(dev); mlx4_free_eq_table(dev); mlx4_close_hca(dev); + if (mlx4_is_mfunc(dev)) + mlx4_multi_func_cleanup(dev); mlx4_cmd_cleanup(dev); if (dev->flags & MLX4_FLAG_MSI_X) pci_disable_msix(pdev); + if (sr_iov && (dev->flags & MLX4_FLAG_SRIOV)) { + mlx4_warn(dev, "Disabling sriov\n"); + pci_disable_sriov(pdev); + } kfree(priv); pci_release_regions(pdev); @@ -1358,18 +1504,31 @@ int mlx4_restart_one(struct pci_dev *pdev) } static DEFINE_PCI_DEVICE_TABLE(mlx4_pci_table) = { - { PCI_VDEVICE(MELLANOX, 0x6340) }, /* MT25408 "Hermon" SDR */ - { PCI_VDEVICE(MELLANOX, 0x634a) }, /* MT25408 "Hermon" DDR */ - { PCI_VDEVICE(MELLANOX, 0x6354) }, /* MT25408 "Hermon" QDR */ - { PCI_VDEVICE(MELLANOX, 0x6732) }, /* MT25408 "Hermon" DDR PCIe gen2 */ - { PCI_VDEVICE(MELLANOX, 0x673c) }, /* MT25408 "Hermon" QDR PCIe gen2 */ - { PCI_VDEVICE(MELLANOX, 0x6368) }, /* MT25408 "Hermon" EN 10GigE */ - { PCI_VDEVICE(MELLANOX, 0x6750) }, /* MT25408 "Hermon" EN 10GigE PCIe gen2 */ - { PCI_VDEVICE(MELLANOX, 0x6372) }, /* MT25458 ConnectX EN 10GBASE-T 10GigE */ - { PCI_VDEVICE(MELLANOX, 0x675a) }, /* MT25458 ConnectX EN 10GBASE-T+Gen2 10GigE */ - { PCI_VDEVICE(MELLANOX, 0x6764) }, /* MT26468 ConnectX EN 10GigE PCIe gen2*/ - { PCI_VDEVICE(MELLANOX, 0x6746) }, /* MT26438 ConnectX EN 40GigE PCIe gen2 5GT/s */ - { PCI_VDEVICE(MELLANOX, 0x676e) }, /* MT26478 ConnectX2 40GigE PCIe gen2 */ + { MLX4_VDEVICE(MELLANOX, 0x6340, 0) }, /* MT25408 "Hermon" SDR */ + { MLX4_VDEVICE(MELLANOX, 0x6341, MLX4_VF) }, /* MT25408 "Hermon" SDR VF */ + { MLX4_VDEVICE(MELLANOX, 0x634a, 0) }, /* MT25408 "Hermon" DDR */ + { MLX4_VDEVICE(MELLANOX, 0x634b, MLX4_VF) }, /* MT25408 "Hermon" DDR VF */ + { MLX4_VDEVICE(MELLANOX, 0x6354, 0) }, /* MT25408 "Hermon" QDR */ + { MLX4_VDEVICE(MELLANOX, 0x6732, 0) }, /* MT25408 "Hermon" DDR PCIe gen2 */ + { MLX4_VDEVICE(MELLANOX, 0x6733, MLX4_VF) }, /* MT25408 "Hermon" DDR PCIe gen2 VF */ + { MLX4_VDEVICE(MELLANOX, 0x673c, 0) }, /* MT25408 "Hermon" QDR PCIe gen2 */ + { MLX4_VDEVICE(MELLANOX, 0x673d, MLX4_VF) }, /* MT25408 "Hermon" QDR PCIe gen2 VF */ + { MLX4_VDEVICE(MELLANOX, 0x6368, 0) }, /* MT25408 "Hermon" EN 10GigE */ + { MLX4_VDEVICE(MELLANOX, 0x6369, MLX4_VF) }, /* MT25408 "Hermon" EN 10GigE VF */ + { MLX4_VDEVICE(MELLANOX, 0x6750, 0) }, /* MT25408 "Hermon" EN 10GigE PCIe gen2 */ + { MLX4_VDEVICE(MELLANOX, 0x6751, MLX4_VF) }, /* MT25408 "Hermon" EN 10GigE PCIe gen2 VF */ + { MLX4_VDEVICE(MELLANOX, 0x6372, 0) }, /* MT25458 ConnectX EN 10GBASE-T 10GigE */ + { MLX4_VDEVICE(MELLANOX, 0x6373, MLX4_VF) }, /* MT25458 ConnectX EN 10GBASE-T 10GigE */ + { MLX4_VDEVICE(MELLANOX, 0x675a, 0) }, /* MT25458 ConnectX EN 10GBASE-T+Gen2 10GigE */ + { MLX4_VDEVICE(MELLANOX, 0x675b, MLX4_VF) }, /* MT25458 ConnectX EN 10GBASE-T+Gen2 10GigE */ + { MLX4_VDEVICE(MELLANOX, 0x6764, 0) }, /* MT26468 ConnectX EN 10GigE PCIe gen2*/ + { MLX4_VDEVICE(MELLANOX, 0x6765, MLX4_VF) }, /* MT26468 ConnectX EN 10GigE PCIe gen2 VF*/ + { MLX4_VDEVICE(MELLANOX, 0x6746, 0) }, /* MT26438 ConnectX VPI PCIe 2.0 5GT/s - IB QDR / 10GigE Virt+ */ + { MLX4_VDEVICE(MELLANOX, 0x6747, MLX4_VF) }, /* MT26438 ConnectX VPI PCIe 2.0 5GT/s - IB QDR / 10GigE Virt+ VF*/ + { MLX4_VDEVICE(MELLANOX, 0x676e, 0) }, /* MT26478 ConnectX EN 40GigE PCIe 2.0 5GT/s */ + { MLX4_VDEVICE(MELLANOX, 0x676f, MLX4_VF) }, /* MT26478 ConnectX EN 40GigE PCIe 2.0 5GT/s VF*/ + { MLX4_VDEVICE(MELLANOX, 0x6778, 0) }, /* MT26488 ConnectX VPI PCIe 2.0 5GT/s - IB DDR / 10GigE Virt+ */ + { MLX4_VDEVICE(MELLANOX, 0x6779, MLX4_VF) }, /* MT26488 ConnectX VPI PCIe 2.0 5GT/s - IB DDR / 10GigE Virt+ VF*/ { 0, } }; diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h index 8ad45f3..5206459 100644 --- a/drivers/net/mlx4/mlx4.h +++ b/drivers/net/mlx4/mlx4.h @@ -144,6 +144,10 @@ extern int mlx4_debug_level; #define MLX4_MAX_NUM_VF 64 #define MLX4_MAX_NUM_SLAVES (MLX4_MAX_NUM_PF + MLX4_MAX_NUM_VF) +#define MLX4_VF (1 << 0) +#define MLX4_VDEVICE(vendor, device, flags) \ + PCI_VDEVICE(vendor, device), (flags) + struct mlx4_bitmap { u32 last; u32 top; -- 1.6.0.2 -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
