mlx4_core: Fix crash on uninitialized priv->cmd.slave_sem

On an SR-IOV master device, __mlx4_init_one() calls mlx4_init_hca()
before mlx4_multi_func_init().  However, for unlucky configurations,
mlx4_init_hca() might call mlx4_SENSE_PORT() (via mlx4_dev_cap()), and
that calls mlx4_cmd_imm() with MLX4_CMD_WRAPPED set.

However, on a multifunction device with MLX4_CMD_WRAPPED, __mlx4_cmd()
calls into mlx4_slave_cmd(), and that immediately tries to do

	down(&priv->cmd.slave_sem);

but priv->cmd.slave_sem isn't initialized until mlx4_multi_func_init()
(which we haven't called yet).  The next thing it tries to do is access
priv->mfunc.vhcr, but that hasn't been allocated yet.

Fix this by moving the initialization of slave_sem and vhcr up into
mlx4_cmd_init(). Also, since slave_sem is really just being used as a
mutex, convert it into a slave_cmd_mutex.

Signed-off-by: Roland Dreier <roland@purestorage.com>

diff --git a/drivers/net/ethernet/mellanox/mlx4/cmd.c b/drivers/net/ethernet/mellanox/mlx4/cmd.c
index 90774b7e47a4b05b8b1a01b7f9d80a791e97fc31..3d1899ff1076e753cfa5a07f7f199a4f01904b2e 100644 (file)
--- a/drivers/net/ethernet/mellanox/mlx4/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c
@@ -395,7 +395,8 @@ static int mlx4_slave_cmd(struct mlx4_dev *dev, u64 in_param, u64 *out_param,
        struct mlx4_vhcr_cmd *vhcr = priv->mfunc.vhcr;
        int ret;
 
-       down(&priv->cmd.slave_sem);
+       mutex_lock(&priv->cmd.slave_cmd_mutex);
+
        vhcr->in_param = cpu_to_be64(in_param);
        vhcr->out_param = out_param ? cpu_to_be64(*out_param) : 0;
        vhcr->in_modifier = cpu_to_be32(in_modifier);
@@ -403,6 +404,7 @@ static int mlx4_slave_cmd(struct mlx4_dev *dev, u64 in_param, u64 *out_param,
        vhcr->token = cpu_to_be16(CMD_POLL_TOKEN);
        vhcr->status = 0;
        vhcr->flags = !!(priv->cmd.use_events) << 6;
+
        if (mlx4_is_master(dev)) {
                ret = mlx4_master_process_vhcr(dev, dev->caps.function, vhcr);
                if (!ret) {
@@ -439,7 +441,8 @@ static int mlx4_slave_cmd(struct mlx4_dev *dev, u64 in_param, u64 *out_param,
                        mlx4_err(dev, "failed execution of VHCR_POST command"
                                 "opcode 0x%x\n", op);
        }
-       up(&priv->cmd.slave_sem);
+
+       mutex_unlock(&priv->cmd.slave_cmd_mutex);
        return ret;
 }
 
@@ -1559,14 +1562,15 @@ static void mlx4_master_do_cmd(struct mlx4_dev *dev, int slave, u8 cmd,
                if ((slave_state[slave].last_cmd != MLX4_COMM_CMD_VHCR_EN) &&
                    (slave_state[slave].last_cmd != MLX4_COMM_CMD_VHCR_POST))
                        goto reset_slave;
-               down(&priv->cmd.slave_sem);
+
+               mutex_lock(&priv->cmd.slave_cmd_mutex);
                if (mlx4_master_process_vhcr(dev, slave, NULL)) {
                        mlx4_err(dev, "Failed processing vhcr for slave:%d,"
                                 " resetting slave.\n", slave);
-                       up(&priv->cmd.slave_sem);
+                       mutex_unlock(&priv->cmd.slave_cmd_mutex);
                        goto reset_slave;
                }
-               up(&priv->cmd.slave_sem);
+               mutex_unlock(&priv->cmd.slave_cmd_mutex);
                break;
        default:
                mlx4_warn(dev, "Bad comm cmd:%d from slave:%d\n", cmd, slave);
@@ -1707,14 +1711,6 @@ int mlx4_multi_func_init(struct mlx4_dev *dev)
        struct mlx4_slave_state *s_state;
        int i, j, err, port;
 
-       priv->mfunc.vhcr = dma_alloc_coherent(&(dev->pdev->dev), PAGE_SIZE,
-                                           &priv->mfunc.vhcr_dma,
-                                           GFP_KERNEL);
-       if (!priv->mfunc.vhcr) {
-               mlx4_err(dev, "Couldn't allocate vhcr.\n");
-               return -ENOMEM;
-       }
-
        if (mlx4_is_master(dev))
                priv->mfunc.comm =
                ioremap(pci_resource_start(dev->pdev, priv->fw.comm_bar) +
@@ -1777,7 +1773,6 @@ int mlx4_multi_func_init(struct mlx4_dev *dev)
                if (mlx4_init_resource_tracker(dev))
                        goto err_thread;
 
-               sema_init(&priv->cmd.slave_sem, 1);
                err = mlx4_ARM_COMM_CHANNEL(dev);
                if (err) {
                        mlx4_err(dev, " Failed to arm comm channel eq: %x\n",
@@ -1791,8 +1786,6 @@ int mlx4_multi_func_init(struct mlx4_dev *dev)
                        mlx4_err(dev, "Couldn't sync toggles\n");
                        goto err_comm;
                }
-
-               sema_init(&priv->cmd.slave_sem, 1);
        }
        return 0;
 
@@ -1822,6 +1815,7 @@ int mlx4_cmd_init(struct mlx4_dev *dev)
        struct mlx4_priv *priv = mlx4_priv(dev);
 
        mutex_init(&priv->cmd.hcr_mutex);
+       mutex_init(&priv->cmd.slave_cmd_mutex);
        sema_init(&priv->cmd.poll_sem, 1);
        priv->cmd.use_events = 0;
        priv->cmd.toggle     = 1;
@@ -1838,14 +1832,30 @@ int mlx4_cmd_init(struct mlx4_dev *dev)
                }
        }
 
+       if (mlx4_is_mfunc(dev)) {
+               priv->mfunc.vhcr = dma_alloc_coherent(&(dev->pdev->dev), PAGE_SIZE,
+                                                     &priv->mfunc.vhcr_dma,
+                                                     GFP_KERNEL);
+               if (!priv->mfunc.vhcr) {
+                       mlx4_err(dev, "Couldn't allocate VHCR.\n");
+                       goto err_hcr;
+               }
+       }
+
        priv->cmd.pool = pci_pool_create("mlx4_cmd", dev->pdev,
                                         MLX4_MAILBOX_SIZE,
                                         MLX4_MAILBOX_SIZE, 0);
        if (!priv->cmd.pool)
-               goto err_hcr;
+               goto err_vhcr;
 
        return 0;
 
+err_vhcr:
+       if (mlx4_is_mfunc(dev))
+               dma_free_coherent(&(dev->pdev->dev), PAGE_SIZE,
+                                 priv->mfunc.vhcr, priv->mfunc.vhcr_dma);
+       priv->mfunc.vhcr = NULL;
+
 err_hcr:
        if (!mlx4_is_slave(dev))
                iounmap(priv->cmd.hcr);
@@ -1868,9 +1878,6 @@ void mlx4_multi_func_cleanup(struct mlx4_dev *dev)
        }
 
        iounmap(priv->mfunc.comm);
-       dma_free_coherent(&(dev->pdev->dev), PAGE_SIZE,
-                    priv->mfunc.vhcr, priv->mfunc.vhcr_dma);
-       priv->mfunc.vhcr = NULL;
 }
 
 void mlx4_cmd_cleanup(struct mlx4_dev *dev)
@@ -1881,6 +1888,10 @@ void mlx4_cmd_cleanup(struct mlx4_dev *dev)
 
        if (!mlx4_is_slave(dev))
                iounmap(priv->cmd.hcr);
+       if (mlx4_is_mfunc(dev))
+               dma_free_coherent(&(dev->pdev->dev), PAGE_SIZE,
+                                 priv->mfunc.vhcr, priv->mfunc.vhcr_dma);
+       priv->mfunc.vhcr = NULL;
 }
 
 /*

diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index 296288c7cb37caae58c5cbe8cb5fe910c54b919c..1e9f816a94dbf643ad5f4cdf3d8a12a63d5f22e7 100644 (file)
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -1159,10 +1159,10 @@ static void mlx4_slave_exit(struct mlx4_dev *dev)
 {
        struct mlx4_priv *priv = mlx4_priv(dev);
 
-       down(&priv->cmd.slave_sem);
+       mutex_lock(&priv->cmd.slave_cmd_mutex);
        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.slave_sem);
+       mutex_unlock(&priv->cmd.slave_cmd_mutex);
 }
 
 static int map_bf_area(struct mlx4_dev *dev)
@@ -1214,7 +1214,7 @@ static int mlx4_init_slave(struct mlx4_dev *dev)
        u32 slave_read;
        u32 cmd_channel_ver;
 
-       down(&priv->cmd.slave_sem);
+       mutex_lock(&priv->cmd.slave_cmd_mutex);
        priv->cmd.max_cmds = 1;
        mlx4_warn(dev, "Sending reset\n");
        ret_from_reset = mlx4_comm_cmd(dev, MLX4_COMM_CMD_RESET, 0,
@@ -1263,12 +1263,13 @@ static int mlx4_init_slave(struct mlx4_dev *dev)
                goto err;
        if (mlx4_comm_cmd(dev, MLX4_COMM_CMD_VHCR_EN, dma, MLX4_COMM_TIME))
                goto err;
-       up(&priv->cmd.slave_sem);
+
+       mutex_unlock(&priv->cmd.slave_cmd_mutex);
        return 0;
 
 err:
        mlx4_comm_cmd(dev, MLX4_COMM_CMD_RESET, 0, 0);
-       up(&priv->cmd.slave_sem);
+       mutex_unlock(&priv->cmd.slave_cmd_mutex);
        return -EIO;
 }
 

diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
index 4b1fbd84a68fe5aa07b57924b4e4e99fa81f4740..71eed05426ec39260f856ce336c9b99fc81f00b8 100644 (file)
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
@@ -513,9 +513,9 @@ struct mlx4_cmd {
        struct pci_pool        *pool;
        void __iomem           *hcr;
        struct mutex            hcr_mutex;
+       struct mutex            slave_cmd_mutex;
        struct semaphore        poll_sem;
        struct semaphore        event_sem;
-       struct semaphore        slave_sem;
        int                     max_cmds;
        spinlock_t              context_lock;
        int                     free_head;