Skip to contentSkip to conversation

IB/mlx5: Clean mr_cache debugfs in case of failure

PATCH In review Leon Romanovsky 2 participants 2 messages May 30, 2017 – Jul 28, 2017 1 message not in the archive

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 IB/mlx5: Clean mr_cache debugfs in case of failure

From: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/[email protected]>

The failure in creation of debugfs entries for mr_cache left entries, which were already created.

It caused to mismatch and misguiding for the end users. The solution is to clean mr_cache debugfs root, so no leftovers will be in the system. In addition, let's document why the error is not needed to be forwarded to user in case of failure.

---
 drivers/infiniband/hw/mlx5/mr.c | 34 ++++++++++++++++++++++------------
 1 file changed, 22 insertions(+), 12 deletions(-)
drivers/infiniband/hw/mlx5/mr.c +22 -12
index 366433f71b58..cc311b287cf5 100644
@@ -582,6 +582,15 @@ static void clean_keys(struct mlx5_ib_dev *dev, int c)
 	}
 }
 
+static void mlx5_mr_cache_debugfs_cleanup(struct mlx5_ib_dev *dev)
+{
+	if (!mlx5_debugfs_root)
+		return;
+
+	debugfs_remove_recursive(dev->cache.root);
+	dev->cache.root = NULL;
+}
+
 static int mlx5_mr_cache_debugfs_init(struct mlx5_ib_dev *dev)
 {
 	struct mlx5_mr_cache *cache = &dev->cache;
@@ -600,38 +609,34 @@ static int mlx5_mr_cache_debugfs_init(struct mlx5_ib_dev *dev)
 		sprintf(ent->name, "%d", ent->order);
 		ent->dir = debugfs_create_dir(ent->name,  cache->root);
 		if (!ent->dir)
-			return -ENOMEM;
+			goto err;
 
 		ent->fsize = debugfs_create_file("size", 0600, ent->dir, ent,
 						 &size_fops);
 		if (!ent->fsize)
-			return -ENOMEM;
+			goto err;
 
 		ent->flimit = debugfs_create_file("limit", 0600, ent->dir, ent,
 						  &limit_fops);
 		if (!ent->flimit)
-			return -ENOMEM;
+			goto err;
 
 		ent->fcur = debugfs_create_u32("cur", 0400, ent->dir,
 					       &ent->cur);
 		if (!ent->fcur)
-			return -ENOMEM;
+			goto err;
 
 		ent->fmiss = debugfs_create_u32("miss", 0600, ent->dir,
 						&ent->miss);
 		if (!ent->fmiss)
-			return -ENOMEM;
+			goto err;
 	}
 
 	return 0;
-}
-
-static void mlx5_mr_cache_debugfs_cleanup(struct mlx5_ib_dev *dev)
-{
-	if (!mlx5_debugfs_root)
-		return;
+err:
+	mlx5_mr_cache_debugfs_cleanup(dev);
 
-	debugfs_remove_recursive(dev->cache.root);
+	return -ENOMEM;
 }
 
 static void delay_time_func(unsigned long ctx)
@@ -692,6 +697,11 @@ int mlx5_mr_cache_init(struct mlx5_ib_dev *dev)
 	if (err)
 		mlx5_ib_warn(dev, "cache debugfs failure\n");
 
+	/*
+	 * We don't want to fail driver if debugfs failed to initialize,
+	 * so we are not forwarding error to the user.
+	 */
+
 	return 0;
 }
 

--
2.12.2

--
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

A message that is not in the archive (sent off-list or not yet received): 20170530064448.4282-1-leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org

#2 lore raw
Quoted 5 lines · Leon Romanovsky

From: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/[email protected]>

The failure in creation of debugfs entries for mr_cache left entries, which were already created.

Quoted 8 lines

It caused to mismatch and misguiding for the end users. The solution is to clean mr_cache debugfs root, so no leftovers will be in the system. In addition, let's document why the error is not needed to be forwarded to user in case of failure.

This was actually submitted to Linus and already accepted in the 4.13- rc cycle, but I missed replying to the message in my linux-rdma folder.
Sorry about that.

Signature · 8 lines
Doug Ledford <[email protected]>
    GPG KeyID: B826A3330E572FDD
    Key fingerprint = AE6B 1BDA 122B 23B4 265B  1274 B826 A333 0E57 2FDD

--
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