From: "Steven Rostedt (Google)" <rost...@goodmis.org>
eventfs uses the tracefs_inode and assumes that it's already initialized
to zero. That is, it doesn't set fields to zero (like ti->private) after
getting its tracefs_inode. This causes bugs due to stale values.
Just initialize the entire structure to zero on allocation so there isn't
any more surprises.
This is a partial fix for accessing ti->private. The assignment still needs
to be made before the dentry is instantiated.
Cc: sta...@vger.kernel.org
Fixes: 5790b1fb3d672 ("eventfs: Remove eventfs_file and just use eventfs_inode")
Reported-by: kernel test robot <oliver.s...@intel.com>
Closes:
https://lore.kernel.org/oe-lkp/202401291043.e62e89dc-oliver.s...@intel.com
Suggested-by: Linus Torvalds <torva...@linux-foundation.org>
Signed-off-by: Steven Rostedt (Google) <rost...@goodmis.org>
---
Changes since v1:
https://lore.kernel.org/all/20240130151737.6e97a...@gandalf.local.home/
- I didn't realize the slab had a constructor which prohibits __GFP_ZERO.
Just use memset() in the constructor. I noticed the WARN_ON_ONCE
that was triggering early in the boot process.
fs/tracefs/inode.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/fs/tracefs/inode.c b/fs/tracefs/inode.c
index e1b172c0e091..636180d45c62 100644
--- a/fs/tracefs/inode.c
+++ b/fs/tracefs/inode.c
@@ -38,8 +38,6 @@ static struct inode *tracefs_alloc_inode(struct super_block
*sb)
if (!ti)
return NULL;
- ti->flags = 0;
-
return &ti->vfs_inode;
}
@@ -779,6 +777,7 @@ static void init_once(void *foo)
{
struct tracefs_inode *ti = (struct tracefs_inode *) foo;
+ memset(ti, 0, sizeof(*ti));
inode_init_once(&ti->vfs_inode);
}
--
2.43.0
