vfs_getxattr_alloc(): don't allocate buf on failure
PATCH 1/1 vfs_getxattr_alloc(): don't allocate buf on failure
Some callers of vfs_getxattr_alloc() assume that on failure the allocated buffer does not need to be freed.
Callers could be fixed, but fixing the semantics of vfs_getxattr_alloc() is simpler and makes sure that this class of bugs does not occur again.
--- fs/xattr.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
fs/xattr.c +4 -1
@@ -383,7 +383,10 @@ vfs_getxattr_alloc(struct user_namespace *mnt_userns, struct dentry *dentry, } error = handler->get (handler, dentry, inode, name, value, error); - *xattr_value = value; +if (error <0 && value != *xattr_value) +kfree (value); +else + *xattr_value = value;return error; }
Signature · 1 line
2.35.3
Al Viro
maintainerQuoted 18 lines · Miklos Szeredi
Some callers of
vfs_getxattr_alloc()assume that on failure the allocated buffer does not need to be freed.Callers could be fixed, but fixing the semantics of
vfs_getxattr_alloc()is simpler and makes sure that this class of bugs does not occur again.Reported-and-tested-by [email protected]Fixes 1601fbad2b14 ("xattr: define vfs_getxattr_alloc and vfs_xattr_cmp")Signed-off-by Miklos Szeredi <[email protected]>--- fs/xattr.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)fs/xattr.c +0
-0
fs/xattr.c vfs_getxattr_alloc(struct user_namespace *mnt_userns, struct dentry *dentry, · line 391
Quoted 10 lines
fs/xattr.c +4
-1@@ -383,7 +383,10 @@ vfs_getxattr_alloc(struct user_namespace *mnt_userns, struct dentry *dentry, } error = handler->get (handler, dentry, inode, name, value, error); - *xattr_value = value; +if (error <0 && value != *xattr_value) +kfree (value); +else + *xattr_value = value;return error; }
Think what happens if it had been called with non-NULL *xattr_value, found that it needed realloc, had krealloc() succeed (and free the original), only to fail in ->get().
Your variant will leave *xattr_value pointing to already freed object, with no way for the caller to tell that from failure before it got to krealloc().
IOW, that's unusable for callers with preallocated buffer - in particular, ones that call that thing in a loop.
Al Viro
maintainerQuoted 34 lines · Al Viro
Quoted 18 lines · Miklos Szeredi
Some callers of
vfs_getxattr_alloc()assume that on failure the allocated buffer does not need to be freed.Callers could be fixed, but fixing the semantics of
vfs_getxattr_alloc()is simpler and makes sure that this class of bugs does not occur again.Reported-and-tested-by [email protected]Fixes 1601fbad2b14 ("xattr: define vfs_getxattr_alloc and vfs_xattr_cmp")Signed-off-by Miklos Szeredi <[email protected]>--- fs/xattr.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)fs/xattr.c +0
-0fs/xattr.c vfs_getxattr_alloc(struct user_namespace *mnt_userns, struct dentry *dentry, · line 391
Quoted 10 lines
fs/xattr.c +4
-1@@ -383,7 +383,10 @@ vfs_getxattr_alloc(struct user_namespace *mnt_userns, struct dentry *dentry, } error = handler->get (handler, dentry, inode, name, value, error); - *xattr_value = value; +if (error <0 && value != *xattr_value) +kfree (value); +else + *xattr_value = value;return error; }Think what happens if it had been called with non-NULL
*xattr_value, found that it needed realloc, hadkrealloc()succeed (and free the original), only to fail in ->get().
Quoted 6 lines
Your variant will leave
*xattr_valuepointing to already freed object, with no way for the caller to tell that from failure before it got tokrealloc().IOW, that's unusable for callers with preallocated buffer - in particular, ones that call that thing in a loop.
FWIW, if we change calling conventions so that in some cases caller need not kfree() whatever's in *xattr_value, about the only variant I see is to have the damn thing freed and replaced with NULL on *all* failure exits. Might or might not make sense, not sure...
Quoted 40 lines · Al Viro
Quoted 38 lines · Al Viro
Quoted 18 lines · Miklos Szeredi
Some callers of
vfs_getxattr_alloc()assume that on failure the allocated buffer does not need to be freed.Callers could be fixed, but fixing the semantics of
vfs_getxattr_alloc()is simpler and makes sure that this class of bugs does not occur again.Reported-and-tested-by [email protected]Fixes 1601fbad2b14 ("xattr: define vfs_getxattr_alloc and vfs_xattr_cmp")Signed-off-by Miklos Szeredi <[email protected]>--- fs/xattr.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)fs/xattr.c +0
-0fs/xattr.c vfs_getxattr_alloc(struct user_namespace *mnt_userns, struct dentry *dentry, · line 391
Quoted 10 lines
fs/xattr.c +4
-1@@ -383,7 +383,10 @@ vfs_getxattr_alloc(struct user_namespace *mnt_userns, struct dentry *dentry, } error = handler->get (handler, dentry, inode, name, value, error); - *xattr_value = value; +if (error <0 && value != *xattr_value) +kfree (value); +else + *xattr_value = value;return error; }Think what happens if it had been called with non-NULL
*xattr_value, found that it needed realloc, hadkrealloc()succeed (and free the original), only to fail in ->get().Your variant will leave
*xattr_valuepointing to already freed object, with no way for the caller to tell that from failure before it got tokrealloc().
Quoted 7 lines
Quoted 2 lines
IOW, that's unusable for callers with preallocated buffer - in particular, ones that call that thing in a loop.
FWIW, if we change calling conventions so that in some cases caller need not
kfree()whatever's in*xattr_value, about the only variant I see is to have the damn thing freed and replaced with NULL on *all* failure exits. Might or might not make sense, not sure...
The minimal semantic change that would fix all buggy callers is to change the non-loop case (i.e. when called only once with a xattr_value pointing to NULL.
In the looping case callers correctly handle the error, since an explicit kfree in the helper is unexpected.
Posted v2.
Thanks,
Miklos