From f9baff509f8a05a79626defdbdf4f4aa4efd373b Mon Sep 17 00:00:00 2001 From: Jack Morgenstein Date: Tue, 13 Dec 2011 04:10:51 +0000 Subject: mlx4_core: Add "native" argument to mlx4_cmd and its callers (where needed) For SRIOV, some Hypervisor commands can be executed directly (native = 1). Others should go through the command wrapper flow (for tracking resource usage, for example, or for changing some HCA configurations that slaves need to be notified of). This patch sets the groundwork for this capability -- adding the correct value of "native" in each case. Note that if SRIOV is not activated, this parameter has no effect. Signed-off-by: Jack Morgenstein Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/qp.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'drivers/net/ethernet/mellanox/mlx4/qp.c') diff --git a/drivers/net/ethernet/mellanox/mlx4/qp.c b/drivers/net/ethernet/mellanox/mlx4/qp.c index 15f870cb2590..e721f4cd34f8 100644 --- a/drivers/net/ethernet/mellanox/mlx4/qp.c +++ b/drivers/net/ethernet/mellanox/mlx4/qp.c @@ -119,7 +119,8 @@ int mlx4_qp_modify(struct mlx4_dev *dev, struct mlx4_mtt *mtt, if (op[cur_state][new_state] == MLX4_CMD_2RST_QP) return mlx4_cmd(dev, 0, qp->qpn, 2, - MLX4_CMD_2RST_QP, MLX4_CMD_TIME_CLASS_A); + MLX4_CMD_2RST_QP, MLX4_CMD_TIME_CLASS_A, + MLX4_CMD_WRAPPED); mailbox = mlx4_alloc_cmd_mailbox(dev); if (IS_ERR(mailbox)) @@ -140,7 +141,8 @@ int mlx4_qp_modify(struct mlx4_dev *dev, struct mlx4_mtt *mtt, ret = mlx4_cmd(dev, mailbox->dma, qp->qpn | (!!sqd_event << 31), new_state == MLX4_QP_STATE_RST ? 2 : 0, - op[cur_state][new_state], MLX4_CMD_TIME_CLASS_C); + op[cur_state][new_state], MLX4_CMD_TIME_CLASS_C, + MLX4_CMD_WRAPPED); mlx4_free_cmd_mailbox(dev, mailbox); return ret; @@ -265,7 +267,7 @@ EXPORT_SYMBOL_GPL(mlx4_qp_free); static int mlx4_CONF_SPECIAL_QP(struct mlx4_dev *dev, u32 base_qpn) { return mlx4_cmd(dev, 0, base_qpn, 0, MLX4_CMD_CONF_SPECIAL_QP, - MLX4_CMD_TIME_CLASS_B); + MLX4_CMD_TIME_CLASS_B, MLX4_CMD_NATIVE); } int mlx4_init_qp_table(struct mlx4_dev *dev) @@ -342,7 +344,8 @@ int mlx4_qp_query(struct mlx4_dev *dev, struct mlx4_qp *qp, return PTR_ERR(mailbox); err = mlx4_cmd_box(dev, 0, mailbox->dma, qp->qpn, 0, - MLX4_CMD_QUERY_QP, MLX4_CMD_TIME_CLASS_A); + MLX4_CMD_QUERY_QP, MLX4_CMD_TIME_CLASS_A, + MLX4_CMD_WRAPPED); if (!err) memcpy(context, mailbox->buf + 8, sizeof *context); -- cgit v1.2.1 From fe9a2603c530fbf1e5d798901cec8d5b79976533 Mon Sep 17 00:00:00 2001 From: Jack Morgenstein Date: Tue, 13 Dec 2011 04:13:22 +0000 Subject: mlx4_core: qp modifications for SRIOV QPs are resources which are allocated and tracked by the PF driver. In multifunction mode, the allocation and icm mapping is done in the resource tracker (later patch in this sequence). To accomplish this, we have "work" functions whose names start with "__", and "request" functions (same name, no __). If we are operating in multifunction mode, the request function actually results in comm-channel commands being sent (ALLOC_RES or FREE_RES). The PF-driver comm-channel handler will ultimately invoke the "work" (__) function and return the result. If we are not in multifunction mode, the "work" handler is invoked immediately. Signed-off-by: Jack Morgenstein Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/qp.c | 235 ++++++++++++++++++++++++-------- 1 file changed, 181 insertions(+), 54 deletions(-) (limited to 'drivers/net/ethernet/mellanox/mlx4/qp.c') diff --git a/drivers/net/ethernet/mellanox/mlx4/qp.c b/drivers/net/ethernet/mellanox/mlx4/qp.c index e721f4cd34f8..d0489740563e 100644 --- a/drivers/net/ethernet/mellanox/mlx4/qp.c +++ b/drivers/net/ethernet/mellanox/mlx4/qp.c @@ -35,6 +35,8 @@ #include #include +#include + #include #include @@ -55,7 +57,7 @@ void mlx4_qp_event(struct mlx4_dev *dev, u32 qpn, int event_type) spin_unlock(&qp_table->lock); if (!qp) { - mlx4_warn(dev, "Async event for bogus QP %08x\n", qpn); + mlx4_dbg(dev, "Async event for none existent QP %08x\n", qpn); return; } @@ -65,10 +67,17 @@ void mlx4_qp_event(struct mlx4_dev *dev, u32 qpn, int event_type) complete(&qp->free); } -int mlx4_qp_modify(struct mlx4_dev *dev, struct mlx4_mtt *mtt, - enum mlx4_qp_state cur_state, enum mlx4_qp_state new_state, - struct mlx4_qp_context *context, enum mlx4_qp_optpar optpar, - int sqd_event, struct mlx4_qp *qp) +static int is_qp0(struct mlx4_dev *dev, struct mlx4_qp *qp) +{ + return qp->qpn >= dev->caps.sqp_start && + qp->qpn <= dev->caps.sqp_start + 1; +} + +static int __mlx4_qp_modify(struct mlx4_dev *dev, struct mlx4_mtt *mtt, + enum mlx4_qp_state cur_state, enum mlx4_qp_state new_state, + struct mlx4_qp_context *context, + enum mlx4_qp_optpar optpar, + int sqd_event, struct mlx4_qp *qp, int native) { static const u16 op[MLX4_QP_NUM_STATE][MLX4_QP_NUM_STATE] = { [MLX4_QP_STATE_RST] = { @@ -110,17 +119,26 @@ int mlx4_qp_modify(struct mlx4_dev *dev, struct mlx4_mtt *mtt, } }; + struct mlx4_priv *priv = mlx4_priv(dev); struct mlx4_cmd_mailbox *mailbox; int ret = 0; + u8 port; if (cur_state >= MLX4_QP_NUM_STATE || new_state >= MLX4_QP_NUM_STATE || !op[cur_state][new_state]) return -EINVAL; - if (op[cur_state][new_state] == MLX4_CMD_2RST_QP) - return mlx4_cmd(dev, 0, qp->qpn, 2, - MLX4_CMD_2RST_QP, MLX4_CMD_TIME_CLASS_A, - MLX4_CMD_WRAPPED); + if (op[cur_state][new_state] == MLX4_CMD_2RST_QP) { + ret = mlx4_cmd(dev, 0, qp->qpn, 2, + MLX4_CMD_2RST_QP, MLX4_CMD_TIME_CLASS_A, native); + if (mlx4_is_master(dev) && cur_state != MLX4_QP_STATE_ERR && + cur_state != MLX4_QP_STATE_RST && + is_qp0(dev, qp)) { + port = (qp->qpn & 1) + 1; + priv->mfunc.master.qp0_state[port].qp0_active = 0; + } + return ret; + } mailbox = mlx4_alloc_cmd_mailbox(dev); if (IS_ERR(mailbox)) @@ -133,108 +151,218 @@ int mlx4_qp_modify(struct mlx4_dev *dev, struct mlx4_mtt *mtt, context->log_page_size = mtt->page_shift - MLX4_ICM_PAGE_SHIFT; } + port = ((context->pri_path.sched_queue >> 6) & 1) + 1; + if (dev->caps.port_type[port] == MLX4_PORT_TYPE_ETH) + context->pri_path.sched_queue = (context->pri_path.sched_queue & + 0xc3); + *(__be32 *) mailbox->buf = cpu_to_be32(optpar); memcpy(mailbox->buf + 8, context, sizeof *context); ((struct mlx4_qp_context *) (mailbox->buf + 8))->local_qpn = cpu_to_be32(qp->qpn); - ret = mlx4_cmd(dev, mailbox->dma, qp->qpn | (!!sqd_event << 31), + ret = mlx4_cmd(dev, mailbox->dma | dev->caps.function, + qp->qpn | (!!sqd_event << 31), new_state == MLX4_QP_STATE_RST ? 2 : 0, - op[cur_state][new_state], MLX4_CMD_TIME_CLASS_C, - MLX4_CMD_WRAPPED); + op[cur_state][new_state], MLX4_CMD_TIME_CLASS_C, native); mlx4_free_cmd_mailbox(dev, mailbox); return ret; } + +int mlx4_qp_modify(struct mlx4_dev *dev, struct mlx4_mtt *mtt, + enum mlx4_qp_state cur_state, enum mlx4_qp_state new_state, + struct mlx4_qp_context *context, + enum mlx4_qp_optpar optpar, + int sqd_event, struct mlx4_qp *qp) +{ + return __mlx4_qp_modify(dev, mtt, cur_state, new_state, context, + optpar, sqd_event, qp, 0); +} EXPORT_SYMBOL_GPL(mlx4_qp_modify); -int mlx4_qp_reserve_range(struct mlx4_dev *dev, int cnt, int align, int *base) +static int __mlx4_qp_reserve_range(struct mlx4_dev *dev, int cnt, int align, + int *base) { struct mlx4_priv *priv = mlx4_priv(dev); struct mlx4_qp_table *qp_table = &priv->qp_table; - int qpn; - qpn = mlx4_bitmap_alloc_range(&qp_table->bitmap, cnt, align); - if (qpn == -1) + *base = mlx4_bitmap_alloc_range(&qp_table->bitmap, cnt, align); + if (*base == -1) return -ENOMEM; - *base = qpn; return 0; } + +int mlx4_qp_reserve_range(struct mlx4_dev *dev, int cnt, int align, int *base) +{ + u64 in_param; + u64 out_param; + int err; + + if (mlx4_is_mfunc(dev)) { + set_param_l(&in_param, cnt); + set_param_h(&in_param, align); + err = mlx4_cmd_imm(dev, in_param, &out_param, + RES_QP, RES_OP_RESERVE, + MLX4_CMD_ALLOC_RES, + MLX4_CMD_TIME_CLASS_A, MLX4_CMD_WRAPPED); + if (err) + return err; + + *base = get_param_l(&out_param); + return 0; + } + return __mlx4_qp_reserve_range(dev, cnt, align, base); +} EXPORT_SYMBOL_GPL(mlx4_qp_reserve_range); -void mlx4_qp_release_range(struct mlx4_dev *dev, int base_qpn, int cnt) +static void __mlx4_qp_release_range(struct mlx4_dev *dev, int base_qpn, int cnt) { struct mlx4_priv *priv = mlx4_priv(dev); struct mlx4_qp_table *qp_table = &priv->qp_table; - if (base_qpn < dev->caps.sqp_start + 8) - return; + if (mlx4_is_qp_reserved(dev, (u32) base_qpn)) + return; mlx4_bitmap_free_range(&qp_table->bitmap, base_qpn, cnt); } + +void mlx4_qp_release_range(struct mlx4_dev *dev, int base_qpn, int cnt) +{ + u64 in_param; + int err; + + if (mlx4_is_mfunc(dev)) { + set_param_l(&in_param, base_qpn); + set_param_h(&in_param, cnt); + err = mlx4_cmd(dev, in_param, RES_QP, RES_OP_RESERVE, + MLX4_CMD_FREE_RES, + MLX4_CMD_TIME_CLASS_A, MLX4_CMD_WRAPPED); + if (err) { + mlx4_warn(dev, "Failed to release qp range" + " base:%d cnt:%d\n", base_qpn, cnt); + } + } else + __mlx4_qp_release_range(dev, base_qpn, cnt); +} EXPORT_SYMBOL_GPL(mlx4_qp_release_range); -int mlx4_qp_alloc(struct mlx4_dev *dev, int qpn, struct mlx4_qp *qp) +static int __mlx4_qp_alloc_icm(struct mlx4_dev *dev, int qpn) { struct mlx4_priv *priv = mlx4_priv(dev); struct mlx4_qp_table *qp_table = &priv->qp_table; int err; - if (!qpn) - return -EINVAL; - - qp->qpn = qpn; - - err = mlx4_table_get(dev, &qp_table->qp_table, qp->qpn); + err = mlx4_table_get(dev, &qp_table->qp_table, qpn); if (err) goto err_out; - err = mlx4_table_get(dev, &qp_table->auxc_table, qp->qpn); + err = mlx4_table_get(dev, &qp_table->auxc_table, qpn); if (err) goto err_put_qp; - err = mlx4_table_get(dev, &qp_table->altc_table, qp->qpn); + err = mlx4_table_get(dev, &qp_table->altc_table, qpn); if (err) goto err_put_auxc; - err = mlx4_table_get(dev, &qp_table->rdmarc_table, qp->qpn); + err = mlx4_table_get(dev, &qp_table->rdmarc_table, qpn); if (err) goto err_put_altc; - err = mlx4_table_get(dev, &qp_table->cmpt_table, qp->qpn); + err = mlx4_table_get(dev, &qp_table->cmpt_table, qpn); if (err) goto err_put_rdmarc; - spin_lock_irq(&qp_table->lock); - err = radix_tree_insert(&dev->qp_table_tree, qp->qpn & (dev->caps.num_qps - 1), qp); - spin_unlock_irq(&qp_table->lock); - if (err) - goto err_put_cmpt; - - atomic_set(&qp->refcount, 1); - init_completion(&qp->free); - return 0; -err_put_cmpt: - mlx4_table_put(dev, &qp_table->cmpt_table, qp->qpn); - err_put_rdmarc: - mlx4_table_put(dev, &qp_table->rdmarc_table, qp->qpn); + mlx4_table_put(dev, &qp_table->rdmarc_table, qpn); err_put_altc: - mlx4_table_put(dev, &qp_table->altc_table, qp->qpn); + mlx4_table_put(dev, &qp_table->altc_table, qpn); err_put_auxc: - mlx4_table_put(dev, &qp_table->auxc_table, qp->qpn); + mlx4_table_put(dev, &qp_table->auxc_table, qpn); err_put_qp: - mlx4_table_put(dev, &qp_table->qp_table, qp->qpn); + mlx4_table_put(dev, &qp_table->qp_table, qpn); err_out: return err; } + +static int mlx4_qp_alloc_icm(struct mlx4_dev *dev, int qpn) +{ + u64 param; + + if (mlx4_is_mfunc(dev)) { + set_param_l(¶m, qpn); + return mlx4_cmd_imm(dev, param, ¶m, RES_QP, RES_OP_MAP_ICM, + MLX4_CMD_ALLOC_RES, MLX4_CMD_TIME_CLASS_A, + MLX4_CMD_WRAPPED); + } + return __mlx4_qp_alloc_icm(dev, qpn); +} + +static void __mlx4_qp_free_icm(struct mlx4_dev *dev, int qpn) +{ + struct mlx4_priv *priv = mlx4_priv(dev); + struct mlx4_qp_table *qp_table = &priv->qp_table; + + mlx4_table_put(dev, &qp_table->cmpt_table, qpn); + mlx4_table_put(dev, &qp_table->rdmarc_table, qpn); + mlx4_table_put(dev, &qp_table->altc_table, qpn); + mlx4_table_put(dev, &qp_table->auxc_table, qpn); + mlx4_table_put(dev, &qp_table->qp_table, qpn); +} + +static void mlx4_qp_free_icm(struct mlx4_dev *dev, int qpn) +{ + u64 in_param; + + if (mlx4_is_mfunc(dev)) { + set_param_l(&in_param, qpn); + if (mlx4_cmd(dev, in_param, RES_QP, RES_OP_MAP_ICM, + MLX4_CMD_FREE_RES, MLX4_CMD_TIME_CLASS_A, + MLX4_CMD_WRAPPED)) + mlx4_warn(dev, "Failed to free icm of qp:%d\n", qpn); + } else + __mlx4_qp_free_icm(dev, qpn); +} + +int mlx4_qp_alloc(struct mlx4_dev *dev, int qpn, struct mlx4_qp *qp) +{ + struct mlx4_priv *priv = mlx4_priv(dev); + struct mlx4_qp_table *qp_table = &priv->qp_table; + int err; + + if (!qpn) + return -EINVAL; + + qp->qpn = qpn; + + err = mlx4_qp_alloc_icm(dev, qpn); + if (err) + return err; + + spin_lock_irq(&qp_table->lock); + err = radix_tree_insert(&dev->qp_table_tree, qp->qpn & + (dev->caps.num_qps - 1), qp); + spin_unlock_irq(&qp_table->lock); + if (err) + goto err_icm; + + atomic_set(&qp->refcount, 1); + init_completion(&qp->free); + + return 0; + +err_icm: + mlx4_qp_free_icm(dev, qpn); + return err; +} + EXPORT_SYMBOL_GPL(mlx4_qp_alloc); void mlx4_qp_remove(struct mlx4_dev *dev, struct mlx4_qp *qp) @@ -250,17 +378,11 @@ EXPORT_SYMBOL_GPL(mlx4_qp_remove); void mlx4_qp_free(struct mlx4_dev *dev, struct mlx4_qp *qp) { - struct mlx4_qp_table *qp_table = &mlx4_priv(dev)->qp_table; - if (atomic_dec_and_test(&qp->refcount)) complete(&qp->free); wait_for_completion(&qp->free); - mlx4_table_put(dev, &qp_table->cmpt_table, qp->qpn); - mlx4_table_put(dev, &qp_table->rdmarc_table, qp->qpn); - mlx4_table_put(dev, &qp_table->altc_table, qp->qpn); - mlx4_table_put(dev, &qp_table->auxc_table, qp->qpn); - mlx4_table_put(dev, &qp_table->qp_table, qp->qpn); + mlx4_qp_free_icm(dev, qp->qpn); } EXPORT_SYMBOL_GPL(mlx4_qp_free); @@ -278,6 +400,8 @@ int mlx4_init_qp_table(struct mlx4_dev *dev) spin_lock_init(&qp_table->lock); INIT_RADIX_TREE(&dev->qp_table_tree, GFP_ATOMIC); + if (mlx4_is_slave(dev)) + return 0; /* * We reserve 2 extra QPs per port for the special QPs. The @@ -329,6 +453,9 @@ int mlx4_init_qp_table(struct mlx4_dev *dev) void mlx4_cleanup_qp_table(struct mlx4_dev *dev) { + if (mlx4_is_slave(dev)) + return; + mlx4_CONF_SPECIAL_QP(dev, 0); mlx4_bitmap_cleanup(&mlx4_priv(dev)->qp_table.bitmap); } -- cgit v1.2.1 From c82e9aa0a8bcf7a1643ccb71678bab57f3cb4bc6 Mon Sep 17 00:00:00 2001 From: Eli Cohen Date: Tue, 13 Dec 2011 04:15:24 +0000 Subject: mlx4_core: resource tracking for HCA resources used by guests The resource tracker is used to track usage of HCA resources by the different guests. Virtual functions (VFs) are attached to guest operating systems but resources are allocated from the same pool and are assigned to VFs. It is essential that hostile/buggy guests not be able to affect the operation of other VFs, possibly attached to other guest OSs since ConnectX firmware is not tolerant to misuse of resources. The resource tracker module associates each resource with a VF and maintains state information for the allocated object. It also defines allowed state transitions and enforces them. Relationships between resources are also referred to. For example, CQs are pointed to by QPs, so it is forbidden to destroy a CQ if a QP refers to it. ICM memory is always accessible through the primary function and hence it is allocated by the owner of the primary function. When a guest dies, an FLR is generated for all the VFs it owns and all the resources it used are freed. The tracked resource types are: QPs, CQs, SRQs, MPTs, MTTs, MACs, RES_EQs, and XRCDNs. Signed-off-by: Eli Cohen Signed-off-by: Jack Morgenstein Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/qp.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'drivers/net/ethernet/mellanox/mlx4/qp.c') diff --git a/drivers/net/ethernet/mellanox/mlx4/qp.c b/drivers/net/ethernet/mellanox/mlx4/qp.c index d0489740563e..6b03ac8b9002 100644 --- a/drivers/net/ethernet/mellanox/mlx4/qp.c +++ b/drivers/net/ethernet/mellanox/mlx4/qp.c @@ -182,7 +182,7 @@ int mlx4_qp_modify(struct mlx4_dev *dev, struct mlx4_mtt *mtt, } EXPORT_SYMBOL_GPL(mlx4_qp_modify); -static int __mlx4_qp_reserve_range(struct mlx4_dev *dev, int cnt, int align, +int __mlx4_qp_reserve_range(struct mlx4_dev *dev, int cnt, int align, int *base) { struct mlx4_priv *priv = mlx4_priv(dev); @@ -218,7 +218,7 @@ int mlx4_qp_reserve_range(struct mlx4_dev *dev, int cnt, int align, int *base) } EXPORT_SYMBOL_GPL(mlx4_qp_reserve_range); -static void __mlx4_qp_release_range(struct mlx4_dev *dev, int base_qpn, int cnt) +void __mlx4_qp_release_range(struct mlx4_dev *dev, int base_qpn, int cnt) { struct mlx4_priv *priv = mlx4_priv(dev); struct mlx4_qp_table *qp_table = &priv->qp_table; @@ -248,7 +248,7 @@ void mlx4_qp_release_range(struct mlx4_dev *dev, int base_qpn, int cnt) } EXPORT_SYMBOL_GPL(mlx4_qp_release_range); -static int __mlx4_qp_alloc_icm(struct mlx4_dev *dev, int qpn) +int __mlx4_qp_alloc_icm(struct mlx4_dev *dev, int qpn) { struct mlx4_priv *priv = mlx4_priv(dev); struct mlx4_qp_table *qp_table = &priv->qp_table; @@ -305,7 +305,7 @@ static int mlx4_qp_alloc_icm(struct mlx4_dev *dev, int qpn) return __mlx4_qp_alloc_icm(dev, qpn); } -static void __mlx4_qp_free_icm(struct mlx4_dev *dev, int qpn) +void __mlx4_qp_free_icm(struct mlx4_dev *dev, int qpn) { struct mlx4_priv *priv = mlx4_priv(dev); struct mlx4_qp_table *qp_table = &priv->qp_table; -- cgit v1.2.1