[PATCH v2] binderfs: rework superblock destruction

Christian Brauner posted 1 patch 3 years, 7 months ago
drivers/android/binderfs.c | 30 +++++++++++++++++-------------
1 file changed, 17 insertions(+), 13 deletions(-)
[PATCH v2] binderfs: rework superblock destruction
Posted by Christian Brauner 3 years, 7 months ago
From: Al Viro <viro@zeniv.linux.org.uk>

So far we relied on
.put_super = binderfs_put_super()
to destroy info we stashed in sb->s_fs_info. This gave us the required ordering
between ->evict_inode() and sb->s_fs_info destruction.

But the current implementation of binderfs_fill_super() has a memory leak in
the rare circumstance that d_make_root() fails because ->put_super() is only
called when sb->s_root is initialized. Fix this by removing ->put_super() and
simply do all that work in binderfs_kill_super().

Reported-by: Dongliang Mu <mudongliangabcd@gmail.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org>
---
 drivers/android/binderfs.c | 30 +++++++++++++++++-------------
 1 file changed, 17 insertions(+), 13 deletions(-)

diff --git a/drivers/android/binderfs.c b/drivers/android/binderfs.c
index 588d753a7a19..927776fdeb1a 100644
--- a/drivers/android/binderfs.c
+++ b/drivers/android/binderfs.c
@@ -340,22 +340,10 @@ static int binderfs_show_options(struct seq_file *seq, struct dentry *root)
 	return 0;
 }
 
-static void binderfs_put_super(struct super_block *sb)
-{
-	struct binderfs_info *info = sb->s_fs_info;
-
-	if (info && info->ipc_ns)
-		put_ipc_ns(info->ipc_ns);
-
-	kfree(info);
-	sb->s_fs_info = NULL;
-}
-
 static const struct super_operations binderfs_super_ops = {
 	.evict_inode    = binderfs_evict_inode,
 	.show_options	= binderfs_show_options,
 	.statfs         = simple_statfs,
-	.put_super	= binderfs_put_super,
 };
 
 static inline bool is_binderfs_control_device(const struct dentry *dentry)
@@ -785,11 +773,27 @@ static int binderfs_init_fs_context(struct fs_context *fc)
 	return 0;
 }
 
+static void binderfs_kill_super(struct super_block *sb)
+{
+	struct binderfs_info *info = sb->s_fs_info;
+
+	/*
+	 * During inode eviction struct binderfs_info is needed.
+	 * So first wipe the super_block then free struct binderfs_info.
+	 */
+	kill_litter_super(sb);
+
+	if (info && info->ipc_ns)
+		put_ipc_ns(info->ipc_ns);
+
+	kfree(info);
+}
+
 static struct file_system_type binder_fs_type = {
 	.name			= "binder",
 	.init_fs_context	= binderfs_init_fs_context,
 	.parameters		= binderfs_fs_parameters,
-	.kill_sb		= kill_litter_super,
+	.kill_sb		= binderfs_kill_super,
 	.fs_flags		= FS_USERNS_MOUNT,
 };
 

base-commit: 1c23f9e627a7b412978b4e852793c5e3c3efc555
-- 
2.34.1
Re: [PATCH v2] binderfs: rework superblock destruction
Posted by Greg Kroah-Hartman 3 years, 7 months ago
On Tue, Aug 23, 2022 at 11:53:39AM +0200, Christian Brauner wrote:
> From: Al Viro <viro@zeniv.linux.org.uk>
> 
> So far we relied on
> .put_super = binderfs_put_super()
> to destroy info we stashed in sb->s_fs_info. This gave us the required ordering
> between ->evict_inode() and sb->s_fs_info destruction.
> 
> But the current implementation of binderfs_fill_super() has a memory leak in
> the rare circumstance that d_make_root() fails because ->put_super() is only
> called when sb->s_root is initialized. Fix this by removing ->put_super() and
> simply do all that work in binderfs_kill_super().
> 
> Reported-by: Dongliang Mu <mudongliangabcd@gmail.com>
> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
> Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org>
> ---
>  drivers/android/binderfs.c | 30 +++++++++++++++++-------------
>  1 file changed, 17 insertions(+), 13 deletions(-)

What changed from v1?
Re: [PATCH v2] binderfs: rework superblock destruction
Posted by Christian Brauner 3 years, 7 months ago
On Tue, Aug 23, 2022 at 12:45:26PM +0200, Greg Kroah-Hartman wrote:
> On Tue, Aug 23, 2022 at 11:53:39AM +0200, Christian Brauner wrote:
> > From: Al Viro <viro@zeniv.linux.org.uk>
> > 
> > So far we relied on
> > .put_super = binderfs_put_super()
> > to destroy info we stashed in sb->s_fs_info. This gave us the required ordering
> > between ->evict_inode() and sb->s_fs_info destruction.
> > 
> > But the current implementation of binderfs_fill_super() has a memory leak in
> > the rare circumstance that d_make_root() fails because ->put_super() is only
> > called when sb->s_root is initialized. Fix this by removing ->put_super() and
> > simply do all that work in binderfs_kill_super().
> > 
> > Reported-by: Dongliang Mu <mudongliangabcd@gmail.com>
> > Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
> > Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org>
> > ---
> >  drivers/android/binderfs.c | 30 +++++++++++++++++-------------
> >  1 file changed, 17 insertions(+), 13 deletions(-)
> 
> What changed from v1?

Ah, forgot to add:

/* v2 */
Christian Brauner (Microsoft) <brauner@kernel.org>:
- Call kill_litter_super() before sb->s_fs_info cleanup.