[patch] Btrfs: kfree()ing ERR_PTRs

[patch] Btrfs: kfree()ing ERR_PTRs

[Dan Carpenter]

The "inherit" in btrfs_ioctl_snap_create_v2() and "vol_args" in
btrfs_ioctl_rm_dev() are ERR_PTRs so we can't call kfree() on them.

These kind of bugs are "One Err Bugs" where there is just one error
label that does everything.  I could set the "inherit = NULL" and keep
the single out label but it ends up being more complicated that way.  It
makes the code simpler to re-order the unwind so it's in the mirror
order of the allocation and introduce some new error labels.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>

diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index de505d5..741b92b 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -1702,7 +1702,7 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file,
 	    ~(BTRFS_SUBVOL_CREATE_ASYNC | BTRFS_SUBVOL_RDONLY |
 	      BTRFS_SUBVOL_QGROUP_INHERIT)) {
 		ret = -EOPNOTSUPP;
-		goto out;
+		goto free_args;
 	}
 
 	if (vol_args->flags & BTRFS_SUBVOL_CREATE_ASYNC)
@@ -1712,27 +1712,31 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file,
 	if (vol_args->flags & BTRFS_SUBVOL_QGROUP_INHERIT) {
 		if (vol_args->size > PAGE_CACHE_SIZE) {
 			ret = -EINVAL;
-			goto out;
+			goto free_args;
 		}
 		inherit = memdup_user(vol_args->qgroup_inherit, vol_args->size);
 		if (IS_ERR(inherit)) {
 			ret = PTR_ERR(inherit);
-			goto out;
+			goto free_args;
 		}
 	}
 
 	ret = btrfs_ioctl_snap_create_transid(file, vol_args->name,
 					      vol_args->fd, subvol, ptr,
 					      readonly, inherit);
+	if (ret)
+		goto free_inherit;
 
-	if (ret = 0 && ptr &&
-	    copy_to_user(arg +
-			 offsetof(struct btrfs_ioctl_vol_args_v2,
-				  transid), ptr, sizeof(*ptr)))
+	if (ptr && copy_to_user(arg +
+				offsetof(struct btrfs_ioctl_vol_args_v2,
+					transid),
+				ptr, sizeof(*ptr)))
 		ret = -EFAULT;
-out:
-	kfree(vol_args);
+
+free_inherit:
 	kfree(inherit);
+free_args:
+	kfree(vol_args);
 	return ret;
 }
 
@@ -2649,7 +2653,7 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
 	vol_args = memdup_user(arg, sizeof(*vol_args));
 	if (IS_ERR(vol_args)) {
 		ret = PTR_ERR(vol_args);
-		goto out;
+		goto err_drop;
 	}
 
 	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
@@ -2667,6 +2671,7 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
 
 out:
 	kfree(vol_args);
+err_drop:
 	mnt_drop_write_file(file);
 	return ret;
 }

[patch] Btrfs: kfree()ing ERR_PTRs

[Dan Carpenter]

The "inherit" in btrfs_ioctl_snap_create_v2() and "vol_args" in
btrfs_ioctl_rm_dev() are ERR_PTRs so we can't call kfree() on them.

These kind of bugs are "One Err Bugs" where there is just one error
label that does everything.  I could set the "inherit = NULL" and keep
the single out label but it ends up being more complicated that way.  It
makes the code simpler to re-order the unwind so it's in the mirror
order of the allocation and introduce some new error labels.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>

diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index de505d5..741b92b 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -1702,7 +1702,7 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file,
 	    ~(BTRFS_SUBVOL_CREATE_ASYNC | BTRFS_SUBVOL_RDONLY |
 	      BTRFS_SUBVOL_QGROUP_INHERIT)) {
 		ret = -EOPNOTSUPP;
-		goto out;
+		goto free_args;
 	}
 
 	if (vol_args->flags & BTRFS_SUBVOL_CREATE_ASYNC)
@@ -1712,27 +1712,31 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file,
 	if (vol_args->flags & BTRFS_SUBVOL_QGROUP_INHERIT) {
 		if (vol_args->size > PAGE_CACHE_SIZE) {
 			ret = -EINVAL;
-			goto out;
+			goto free_args;
 		}
 		inherit = memdup_user(vol_args->qgroup_inherit, vol_args->size);
 		if (IS_ERR(inherit)) {
 			ret = PTR_ERR(inherit);
-			goto out;
+			goto free_args;
 		}
 	}
 
 	ret = btrfs_ioctl_snap_create_transid(file, vol_args->name,
 					      vol_args->fd, subvol, ptr,
 					      readonly, inherit);
+	if (ret)
+		goto free_inherit;
 
-	if (ret == 0 && ptr &&
-	    copy_to_user(arg +
-			 offsetof(struct btrfs_ioctl_vol_args_v2,
-				  transid), ptr, sizeof(*ptr)))
+	if (ptr && copy_to_user(arg +
+				offsetof(struct btrfs_ioctl_vol_args_v2,
+					transid),
+				ptr, sizeof(*ptr)))
 		ret = -EFAULT;
-out:
-	kfree(vol_args);
+
+free_inherit:
 	kfree(inherit);
+free_args:
+	kfree(vol_args);
 	return ret;
 }
 
@@ -2649,7 +2653,7 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
 	vol_args = memdup_user(arg, sizeof(*vol_args));
 	if (IS_ERR(vol_args)) {
 		ret = PTR_ERR(vol_args);
-		goto out;
+		goto err_drop;
 	}
 
 	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
@@ -2667,6 +2671,7 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
 
 out:
 	kfree(vol_args);
+err_drop:
 	mnt_drop_write_file(file);
 	return ret;
 }

Re: [patch] Btrfs: kfree()ing ERR_PTRs

[walter harms]

Am 04.09.2014 13:09, schrieb Dan Carpenter:
> The "inherit" in btrfs_ioctl_snap_create_v2() and "vol_args" in
> btrfs_ioctl_rm_dev() are ERR_PTRs so we can't call kfree() on them.
> 
> These kind of bugs are "One Err Bugs" where there is just one error
> label that does everything.  I could set the "inherit = NULL" and keep
> the single out label but it ends up being more complicated that way.  It
> makes the code simpler to re-order the unwind so it's in the mirror
> order of the allocation and introduce some new error labels.
> 
> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
> 
> diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
> index de505d5..741b92b 100644
> --- a/fs/btrfs/ioctl.c
> +++ b/fs/btrfs/ioctl.c
> @@ -1702,7 +1702,7 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file,
>  	    ~(BTRFS_SUBVOL_CREATE_ASYNC | BTRFS_SUBVOL_RDONLY |
>  	      BTRFS_SUBVOL_QGROUP_INHERIT)) {
>  		ret = -EOPNOTSUPP;
> -		goto out;
> +		goto free_args;
>  	}
>  
>  	if (vol_args->flags & BTRFS_SUBVOL_CREATE_ASYNC)
> @@ -1712,27 +1712,31 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file,
>  	if (vol_args->flags & BTRFS_SUBVOL_QGROUP_INHERIT) {
>  		if (vol_args->size > PAGE_CACHE_SIZE) {
>  			ret = -EINVAL;
> -			goto out;
> +			goto free_args;
>  		}
>  		inherit = memdup_user(vol_args->qgroup_inherit, vol_args->size);
>  		if (IS_ERR(inherit)) {
>  			ret = PTR_ERR(inherit);
> -			goto out;
> +			goto free_args;
>  		}
>  	}
>  
>  	ret = btrfs_ioctl_snap_create_transid(file, vol_args->name,
>  					      vol_args->fd, subvol, ptr,
>  					      readonly, inherit);
> +	if (ret)
> +		goto free_inherit;
>  
> -	if (ret = 0 && ptr &&
> -	    copy_to_user(arg +
> -			 offsetof(struct btrfs_ioctl_vol_args_v2,
> -				  transid), ptr, sizeof(*ptr)))
> +	if (ptr && copy_to_user(arg +
> +				offsetof(struct btrfs_ioctl_vol_args_v2,
> +					transid),
> +				ptr, sizeof(*ptr)))
>  		ret = -EFAULT;


this is hard to read.
perhaps it would help to move ptr into the other other

if (ret || !ptr )
  goto free_inherit;

both are modified by the call btrfs_ioctl_snap_create_transid().

@Maintainer:
can this really happen ?

More over i would argue that it would nice to rename
ptr in to transid_ptr or so, this is what the code is all about
and it is not obvious for me.


re,
 wh

> -out:
> -	kfree(vol_args);
> +
> +free_inherit:
>  	kfree(inherit);
> +free_args:
> +	kfree(vol_args);
>  	return ret;
>  }
>  
> @@ -2649,7 +2653,7 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
>  	vol_args = memdup_user(arg, sizeof(*vol_args));
>  	if (IS_ERR(vol_args)) {
>  		ret = PTR_ERR(vol_args);
> -		goto out;
> +		goto err_drop;
>  	}
>  
>  	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
> @@ -2667,6 +2671,7 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
>  
>  out:
>  	kfree(vol_args);
> +err_drop:
>  	mnt_drop_write_file(file);
>  	return ret;
>  }
> --
> To unsubscribe from this list: send the line "unsubscribe kernel-janitors" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

Re: [patch] Btrfs: kfree()ing ERR_PTRs

[walter harms]

Am 04.09.2014 13:09, schrieb Dan Carpenter:
> The "inherit" in btrfs_ioctl_snap_create_v2() and "vol_args" in
> btrfs_ioctl_rm_dev() are ERR_PTRs so we can't call kfree() on them.
> 
> These kind of bugs are "One Err Bugs" where there is just one error
> label that does everything.  I could set the "inherit = NULL" and keep
> the single out label but it ends up being more complicated that way.  It
> makes the code simpler to re-order the unwind so it's in the mirror
> order of the allocation and introduce some new error labels.
> 
> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
> 
> diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
> index de505d5..741b92b 100644
> --- a/fs/btrfs/ioctl.c
> +++ b/fs/btrfs/ioctl.c
> @@ -1702,7 +1702,7 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file,
>  	    ~(BTRFS_SUBVOL_CREATE_ASYNC | BTRFS_SUBVOL_RDONLY |
>  	      BTRFS_SUBVOL_QGROUP_INHERIT)) {
>  		ret = -EOPNOTSUPP;
> -		goto out;
> +		goto free_args;
>  	}
>  
>  	if (vol_args->flags & BTRFS_SUBVOL_CREATE_ASYNC)
> @@ -1712,27 +1712,31 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file,
>  	if (vol_args->flags & BTRFS_SUBVOL_QGROUP_INHERIT) {
>  		if (vol_args->size > PAGE_CACHE_SIZE) {
>  			ret = -EINVAL;
> -			goto out;
> +			goto free_args;
>  		}
>  		inherit = memdup_user(vol_args->qgroup_inherit, vol_args->size);
>  		if (IS_ERR(inherit)) {
>  			ret = PTR_ERR(inherit);
> -			goto out;
> +			goto free_args;
>  		}
>  	}
>  
>  	ret = btrfs_ioctl_snap_create_transid(file, vol_args->name,
>  					      vol_args->fd, subvol, ptr,
>  					      readonly, inherit);
> +	if (ret)
> +		goto free_inherit;
>  
> -	if (ret == 0 && ptr &&
> -	    copy_to_user(arg +
> -			 offsetof(struct btrfs_ioctl_vol_args_v2,
> -				  transid), ptr, sizeof(*ptr)))
> +	if (ptr && copy_to_user(arg +
> +				offsetof(struct btrfs_ioctl_vol_args_v2,
> +					transid),
> +				ptr, sizeof(*ptr)))
>  		ret = -EFAULT;


this is hard to read.
perhaps it would help to move ptr into the other other

if (ret || !ptr )
  goto free_inherit;

both are modified by the call btrfs_ioctl_snap_create_transid().

@Maintainer:
can this really happen ?

More over i would argue that it would nice to rename
ptr in to transid_ptr or so, this is what the code is all about
and it is not obvious for me.


re,
 wh

> -out:
> -	kfree(vol_args);
> +
> +free_inherit:
>  	kfree(inherit);
> +free_args:
> +	kfree(vol_args);
>  	return ret;
>  }
>  
> @@ -2649,7 +2653,7 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
>  	vol_args = memdup_user(arg, sizeof(*vol_args));
>  	if (IS_ERR(vol_args)) {
>  		ret = PTR_ERR(vol_args);
> -		goto out;
> +		goto err_drop;
>  	}
>  
>  	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
> @@ -2667,6 +2671,7 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
>  
>  out:
>  	kfree(vol_args);
> +err_drop:
>  	mnt_drop_write_file(file);
>  	return ret;
>  }
> --
> To unsubscribe from this list: send the line "unsubscribe kernel-janitors" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

Re: [patch] Btrfs: kfree()ing ERR_PTRs

[Dan Carpenter]

On Thu, Sep 04, 2014 at 02:11:00PM +0200, walter harms wrote:
> >  	ret = btrfs_ioctl_snap_create_transid(file, vol_args->name,
> >  					      vol_args->fd, subvol, ptr,
> >  					      readonly, inherit);
> > +	if (ret)
> > +		goto free_inherit;
> >  
> > -	if (ret = 0 && ptr &&
> > -	    copy_to_user(arg +
> > -			 offsetof(struct btrfs_ioctl_vol_args_v2,
> > -				  transid), ptr, sizeof(*ptr)))
> > +	if (ptr && copy_to_user(arg +
> > +				offsetof(struct btrfs_ioctl_vol_args_v2,
> > +					transid),
> > +				ptr, sizeof(*ptr)))
> >  		ret = -EFAULT;
> 
> 
> this is hard to read.
> perhaps it would help to move ptr into the other other
> 
> if (ret || !ptr )
>   goto free_inherit;

I don't think that's an improvement.

> 
> both are modified by the call btrfs_ioctl_snap_create_transid().
> 

I think you are mixing up ptr and *ptr so that's why you're confused.

regards,
dan carpenter

Re: [patch] Btrfs: kfree()ing ERR_PTRs

[Dan Carpenter]

On Thu, Sep 04, 2014 at 02:11:00PM +0200, walter harms wrote:
> >  	ret = btrfs_ioctl_snap_create_transid(file, vol_args->name,
> >  					      vol_args->fd, subvol, ptr,
> >  					      readonly, inherit);
> > +	if (ret)
> > +		goto free_inherit;
> >  
> > -	if (ret == 0 && ptr &&
> > -	    copy_to_user(arg +
> > -			 offsetof(struct btrfs_ioctl_vol_args_v2,
> > -				  transid), ptr, sizeof(*ptr)))
> > +	if (ptr && copy_to_user(arg +
> > +				offsetof(struct btrfs_ioctl_vol_args_v2,
> > +					transid),
> > +				ptr, sizeof(*ptr)))
> >  		ret = -EFAULT;
> 
> 
> this is hard to read.
> perhaps it would help to move ptr into the other other
> 
> if (ret || !ptr )
>   goto free_inherit;

I don't think that's an improvement.

> 
> both are modified by the call btrfs_ioctl_snap_create_transid().
> 

I think you are mixing up ptr and *ptr so that's why you're confused.

regards,
dan carpenter

Re: [patch] Btrfs: kfree()ing ERR_PTRs

[Chris Mason]

On 09/04/2014 07:09 AM, Dan Carpenter wrote:
> The "inherit" in btrfs_ioctl_snap_create_v2() and "vol_args" in
> btrfs_ioctl_rm_dev() are ERR_PTRs so we can't call kfree() on them.
> 
> These kind of bugs are "One Err Bugs" where there is just one error
> label that does everything.  I could set the "inherit = NULL" and keep
> the single out label but it ends up being more complicated that way.  It
> makes the code simpler to re-order the unwind so it's in the mirror
> order of the allocation and introduce some new error labels.

Thanks Dan, agreed on the unwinding.

-chris

Re: [patch] Btrfs: kfree()ing ERR_PTRs

[Chris Mason]

On 09/04/2014 07:09 AM, Dan Carpenter wrote:
> The "inherit" in btrfs_ioctl_snap_create_v2() and "vol_args" in
> btrfs_ioctl_rm_dev() are ERR_PTRs so we can't call kfree() on them.
> 
> These kind of bugs are "One Err Bugs" where there is just one error
> label that does everything.  I could set the "inherit = NULL" and keep
> the single out label but it ends up being more complicated that way.  It
> makes the code simpler to re-order the unwind so it's in the mirror
> order of the allocation and introduce some new error labels.

Thanks Dan, agreed on the unwinding.

-chris