zswap: ignore debugfs_create_dir() return value

PATCH In review Greg Kroah-Hartman 2 participants 4 messages Jan 22, 2019 – Jan 31, 2019

Subsystems (2)

Memory management zswap compressed swap caching

lore

Share & tools

lore

Greg Kroah-Hartman

author

2019-01-22 15:23 UTC #0 lore raw

PATCH 1/1 zswap: ignore debugfs_create_dir() return value

When calling debugfs functions, there is no need to ever check the return value. The function can work or not, but the code logic should never do something different based on this.

Cc Seth Jennings <[email protected]>

Cc Dan Streetman <[email protected]>

Cc [email protected]

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 mm/zswap.c | 2 --
 1 file changed, 2 deletions(-)

mm/zswap.c +0 -2

index a4e4d36ec085..f583d08f6e24 100644
@@ -1262,8 +1262,6 @@ static int __init zswap_debugfs_init(void)
 		return -ENODEV;
 
 	zswap_debugfs_root = debugfs_create_dir("zswap", NULL);
-	if (!zswap_debugfs_root)
-		return -ENOMEM;
 
 	debugfs_create_u64("pool_limit_hit", 0444,
 			   zswap_debugfs_root, &zswap_pool_limit_hit);

Signature · 1 line

2.20.1

Dan Streetman

2019-01-29 19:47 UTC #1 lore raw

Quoted 18 lines · Greg Kroah-Hartman

When calling debugfs functions, there is no need to ever check the return value. The function can work or not, but the code logic should never do something different based on this.
Cc Seth Jennings <[email protected]>
Cc Dan Streetman <[email protected]>
Cc [email protected]
Signed-off-by Greg Kroah-Hartman <[email protected]>
---
 mm/zswap.c | 2 --
 1 file changed, 2 deletions(-)
mm/zswap.c +0 -0
index a4e4d36ec085..f583d08f6e24 100644

mm/zswap.c static int __init zswap_debugfs_init(void) · line 1267

Quoted 8 lines

mm/zswap.c +0 -2

@@ -1262,8 +1262,6 @@ static int __init zswap_debugfs_init(void)
                return -ENODEV;
 
        zswap_debugfs_root = debugfs_create_dir("zswap", NULL);
-       if (!zswap_debugfs_root)
-               return -ENOMEM;
 
        debugfs_create_u64("pool_limit_hit", 0444,
                           zswap_debugfs_root, &zswap_pool_limit_hit);

wait, so if i'm reading the code right, in the case where debugfs_create_dir() returns NULL, that will then be passed along to debugfs_create_u64() as its parent directory - and the debugfs nodes will then get created in the root debugfs directory. That's not what we want to happen...

Quoted 3 lines

--
2.20.1

Greg Kroah-Hartman

author

2019-01-29 20:33 UTC #2 lore raw

Quoted 26 lines · Dan Streetman

Quoted 24 lines · Greg Kroah-Hartman
When calling debugfs functions, there is no need to ever check the return value. The function can work or not, but the code logic should never do something different based on this.
Cc Seth Jennings <[email protected]>
Cc Dan Streetman <[email protected]>
Cc [email protected]
Signed-off-by Greg Kroah-Hartman <[email protected]>
---
 mm/zswap.c | 2 --
 1 file changed, 2 deletions(-)
mm/zswap.c +0 -2
index a4e4d36ec085..f583d08f6e24 100644
@@ -1262,8 +1262,6 @@ static int __init zswap_debugfs_init(void)
                return -ENODEV;
 
        zswap_debugfs_root = debugfs_create_dir("zswap", NULL);
-       if (!zswap_debugfs_root)
-               return -ENOMEM;
 

Quoted 8 lines

mm/zswap.c static int __init zswap_debugfs_init(void) · line 1267
Quoted 2 lines
mm/zswap.c +0 -0
        debugfs_create_u64("pool_limit_hit", 0444,
                           zswap_debugfs_root, &zswap_pool_limit_hit);
wait, so if i'm reading the code right, in the case where debugfs_create_dir() returns NULL, that will then be passed along to debugfs_create_u64() as its parent directory - and the debugfs nodes will then get created in the root debugfs directory. That's not what we want to happen...

True, but that is such a rare thing to ever happen (hint, you have to be out of memory), that it's not really a bad thing. But, you are not the first to mention this, which is why this patch is on its way to Linus for 5.0-final:

	https://lore.kernel.org/lkml/[email protected]/

thanks,

greg k-h

Dan Streetman

Acked-by

2019-01-31 16:09 UTC #3 lore raw

Quoted 37 lines · Greg Kroah-Hartman

Quoted 26 lines · Dan Streetman
Quoted 24 lines · Greg Kroah-Hartman
When calling debugfs functions, there is no need to ever check the return value. The function can work or not, but the code logic should never do something different based on this.
Cc Seth Jennings <[email protected]>
Cc Dan Streetman <[email protected]>
Cc [email protected]
Signed-off-by Greg Kroah-Hartman <[email protected]>
---
 mm/zswap.c | 2 --
 1 file changed, 2 deletions(-)
mm/zswap.c +0 -2
index a4e4d36ec085..f583d08f6e24 100644
@@ -1262,8 +1262,6 @@ static int __init zswap_debugfs_init(void)
                return -ENODEV;
 
        zswap_debugfs_root = debugfs_create_dir("zswap", NULL);
-       if (!zswap_debugfs_root)
-               return -ENOMEM;
 
Quoted 8 lines
mm/zswap.c static int __init zswap_debugfs_init(void) · line 1267
Quoted 2 lines
mm/zswap.c +0 -0
        debugfs_create_u64("pool_limit_hit", 0444,
                           zswap_debugfs_root, &zswap_pool_limit_hit);
wait, so if i'm reading the code right, in the case where debugfs_create_dir() returns NULL, that will then be passed along to debugfs_create_u64() as its parent directory - and the debugfs nodes will then get created in the root debugfs directory. That's not what we want to happen...

Quoted 5 lines

True, but that is such a rare thing to ever happen (hint, you have to be out of memory), that it's not really a bad thing. But, you are not the first to mention this, which is why this patch is on its way to Linus for 5.0-final:
        https://lore.kernel.org/lkml/[email protected]/

Ah! Great, in that case then definitely

Acked-by Dan Streetman <[email protected]>

Quoted 4 lines

thanks,
greg k-h