mm: do file ownership checks with the proper mount idmap

[ Upstream commit e187bc02f8 ]

Ever since idmapped mounts were introduced, inode ownership checks (for
side-channel protection) in mincore() and madvise(MADV_PAGEOUT) were done
against the nop_mnt_idmap, which completely ignores the file's mount's
idmap.  This results in odd edgecases like:

1) mount/bind-mount with an idmap userA:userB:1
2) userB runs an owner_or_capable() check on file that is owned by userA
on-disk/in-memory, but owned by userB after idmap translation
3) owner_or_capable() mysteriously fails as the correct idmap wasn't supplied

In the case of mincore/madvise MADV_PAGEOUT, this is usually benign,
because file_permission(file, MAY_WRITE) will probably succeed, as it uses
the proper idmap internally, but it does not need to be the case on e.g a
0444 file where even the owner itself doesn't have permissions to write to
it.

Since this is clearly not trivial to get right, introduce a
file_owner_or_capable() that can carry the correct semantics, and switch
the various users in mm to it.

The issue was found by manual code inspection & an off-list discussion
with Jan Kara.

Link: https://lore.kernel.org/20260625153853.913949-1-pfalcato@suse.de
Fixes: 9caccd4154 ("fs: introduce MOUNT_ATTR_IDMAP")
Signed-off-by: Pedro Falcato <pfalcato@suse.de>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Christian Brauner (Amutable) <brauner@kernel.org>
Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Jann Horn <jannh@google.com>
Cc: Liam R. Howlett <liam@infradead.org>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Vlastimil Babka <vbabka@kernel.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
[ dropped const from file_owner_or_capable()'s parameter since 6.6's file_mnt_idmap() takes a non-const struct file * ]
Signed-off-by: Sasha Levin <sashal@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

include/linux/fs.h

@@ -2481,6 +2481,11 @@ static inline struct mnt_idmap *file_mnt_idmap(struct file *file)
 	return mnt_idmap(file->f_path.mnt);
 }
 
+static inline bool file_owner_or_capable(struct file *file)
+{
+	return inode_owner_or_capable(file_mnt_idmap(file), file_inode(file));
+}
+
 /**
  * is_idmapped_mnt - check whether a mount is mapped
  * @mnt: the mount to check

mm/filemap.c

@@ -4314,7 +4314,7 @@ static inline bool can_do_cachestat(struct file *f)
 {
 	if (f->f_mode & FMODE_WRITE)
 		return true;
-	if (inode_owner_or_capable(file_mnt_idmap(f), file_inode(f)))
+	if (file_owner_or_capable(f))
 		return true;
 	return file_permission(f, MAY_WRITE) == 0;
 }

mm/madvise.c

@@ -334,8 +334,7 @@ static inline bool can_do_file_pageout(struct vm_area_struct *vma)
 	 * otherwise we'd be including shared non-exclusive mappings, which
 	 * opens a side channel.
 	 */
-	return inode_owner_or_capable(&nop_mnt_idmap,
-				      file_inode(vma->vm_file)) ||
+	return file_owner_or_capable(vma->vm_file) ||
 	       file_permission(vma->vm_file, MAY_WRITE) == 0;
 }
 

mm/mincore.c

@@ -167,8 +167,7 @@ static inline bool can_do_mincore(struct vm_area_struct *vma)
 	 * for writing; otherwise we'd be including shared non-exclusive
 	 * mappings, which opens a side channel.
 	 */
-	return inode_owner_or_capable(&nop_mnt_idmap,
-				      file_inode(vma->vm_file)) ||
+	return file_owner_or_capable(vma->vm_file) ||
 	       file_permission(vma->vm_file, MAY_WRITE) == 0;
 }