net/mlx5: Fix use after free in mlx5_health_wait_pci_up

The device health recovery flow calls mlx5_health_wait_pci_up() which
queries the device for FW_RESET timeout after freeing the device
timeouts structure on mlx5_function_teardown(). Fix this bug by moving
timeouts structure init/cleanup to the device's init/uninit phases.
Since it is necessary to reset default software timeouts on function
reload, extract setting of defaults values from mlx5_tout_init() and
call mlx5_tout_set_def_val() directly from mlx5_function_setup().

Fixes: 5945e1ad ("net/mlx5: Read timeout values from init segment")
Reported by: Niklas Schnelle <schnelle@linux.ibm.com>
Signed-off-by: Amir Tzin <amirtz@nvidia.com>
Signed-off-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

drivers/net/ethernet/mellanox/mlx5/core/lib/tout.c

@@ -31,11 +31,11 @@ static void tout_set(struct mlx5_core_dev *dev, u64 val, enum mlx5_timeouts_type
 	dev->timeouts->to[type] = val;
 }
 
-static void tout_set_def_val(struct mlx5_core_dev *dev)
+void mlx5_tout_set_def_val(struct mlx5_core_dev *dev)
 {
 	int i;
 
-	for (i = MLX5_TO_FW_PRE_INIT_TIMEOUT_MS; i < MAX_TIMEOUT_TYPES; i++)
+	for (i = 0; i < MAX_TIMEOUT_TYPES; i++)
 		tout_set(dev, tout_def_sw_val[i], i);
 }
 
@@ -45,7 +45,6 @@ int mlx5_tout_init(struct mlx5_core_dev *dev)
 	if (!dev->timeouts)
 		return -ENOMEM;
 
-	tout_set_def_val(dev);
 	return 0;
 }
 

drivers/net/ethernet/mellanox/mlx5/core/lib/tout.h

@@ -34,6 +34,7 @@ int mlx5_tout_init(struct mlx5_core_dev *dev);
 void mlx5_tout_cleanup(struct mlx5_core_dev *dev);
 void mlx5_tout_query_iseg(struct mlx5_core_dev *dev);
 int mlx5_tout_query_dtor(struct mlx5_core_dev *dev);
+void mlx5_tout_set_def_val(struct mlx5_core_dev *dev);
 u64 _mlx5_tout_ms(struct mlx5_core_dev *dev, enum mlx5_timeouts_types type);
 
 #define mlx5_tout_ms(dev, type) _mlx5_tout_ms(dev, MLX5_TO_##type##_MS)

drivers/net/ethernet/mellanox/mlx5/core/main.c

@@ -992,11 +992,7 @@ static int mlx5_function_setup(struct mlx5_core_dev *dev, bool boot)
 	if (mlx5_core_is_pf(dev))
 		pcie_print_link_status(dev->pdev);
 
-	err = mlx5_tout_init(dev);
-	if (err) {
-		mlx5_core_err(dev, "Failed initializing timeouts, aborting\n");
-		return err;
-	}
+	mlx5_tout_set_def_val(dev);
 
 	/* wait for firmware to accept initialization segments configurations
 	 */
@@ -1005,13 +1001,13 @@ static int mlx5_function_setup(struct mlx5_core_dev *dev, bool boot)
 	if (err) {
 		mlx5_core_err(dev, "Firmware over %llu MS in pre-initializing state, aborting\n",
 			      mlx5_tout_ms(dev, FW_PRE_INIT_TIMEOUT));
-		goto err_tout_cleanup;
+		return err;
 	}
 
 	err = mlx5_cmd_init(dev);
 	if (err) {
 		mlx5_core_err(dev, "Failed initializing command interface, aborting\n");
-		goto err_tout_cleanup;
+		return err;
 	}
 
 	mlx5_tout_query_iseg(dev);
@@ -1094,8 +1090,6 @@ static int mlx5_function_setup(struct mlx5_core_dev *dev, bool boot)
 err_cmd_cleanup:
 	mlx5_cmd_set_state(dev, MLX5_CMDIF_STATE_DOWN);
 	mlx5_cmd_cleanup(dev);
-err_tout_cleanup:
-	mlx5_tout_cleanup(dev);
 
 	return err;
 }
@@ -1114,7 +1108,6 @@ static int mlx5_function_teardown(struct mlx5_core_dev *dev, bool boot)
 	mlx5_core_disable_hca(dev, 0);
 	mlx5_cmd_set_state(dev, MLX5_CMDIF_STATE_DOWN);
 	mlx5_cmd_cleanup(dev);
-	mlx5_tout_cleanup(dev);
 
 	return 0;
 }
@@ -1476,6 +1469,12 @@ int mlx5_mdev_init(struct mlx5_core_dev *dev, int profile_idx)
 					    mlx5_debugfs_root);
 	INIT_LIST_HEAD(&priv->traps);
 
+	err = mlx5_tout_init(dev);
+	if (err) {
+		mlx5_core_err(dev, "Failed initializing timeouts, aborting\n");
+		goto err_timeout_init;
+	}
+
 	err = mlx5_health_init(dev);
 	if (err)
 		goto err_health_init;
@@ -1501,6 +1500,8 @@ int mlx5_mdev_init(struct mlx5_core_dev *dev, int profile_idx)
 err_pagealloc_init:
 	mlx5_health_cleanup(dev);
 err_health_init:
+	mlx5_tout_cleanup(dev);
+err_timeout_init:
 	debugfs_remove(dev->priv.dbg_root);
 	mutex_destroy(&priv->pgdir_mutex);
 	mutex_destroy(&priv->alloc_mutex);
@@ -1518,6 +1519,7 @@ void mlx5_mdev_uninit(struct mlx5_core_dev *dev)
 	mlx5_adev_cleanup(dev);
 	mlx5_pagealloc_cleanup(dev);
 	mlx5_health_cleanup(dev);
+	mlx5_tout_cleanup(dev);
 	debugfs_remove_recursive(dev->priv.dbg_root);
 	mutex_destroy(&priv->pgdir_mutex);
 	mutex_destroy(&priv->alloc_mutex);