rbd: fix cleanup in rbd_add()

Bjorn Helgaas pointed out that a recent commit introduced a
use-after-free condition in an error path for rbd_add().
He correctly stated:

    I think b536f69a3a "rbd: set up devices only for mapped images"
    introduced a use-after-free error in rbd_add():
	...
    If rbd_dev_device_setup() returns an error, we call
    rbd_dev_image_release(), which ultimately kfrees rbd_dev.
    Then we call rbd_dev_destroy(), which references fields in
    the already-freed rbd_dev struct before kfreeing it again.

The simple fix is to return the error code after the call to
rbd_dev_image_release().

Closer examination revealed that there's no need to clean up
rbd_opts in that function, so fix that too.

Update some other comments that have also become out of date.

Reported-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Alex Elder <elder@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
This commit is contained in:
Alex Elder 2013-05-13 20:35:37 -05:00
parent 7262cfca43
commit 3abef3b358
1 changed files with 14 additions and 9 deletions

View File

@ -4700,8 +4700,10 @@ out:
return ret;
}
/* Undo whatever state changes are made by v1 or v2 image probe */
/*
* Undo whatever state changes are made by v1 or v2 header info
* call.
*/
static void rbd_dev_unprobe(struct rbd_device *rbd_dev)
{
struct rbd_image_header *header;
@ -4905,9 +4907,10 @@ static int rbd_dev_image_probe(struct rbd_device *rbd_dev, bool mapping)
int tmp;
/*
* Get the id from the image id object. If it's not a
* format 2 image, we'll get ENOENT back, and we'll assume
* it's a format 1 image.
* Get the id from the image id object. Unless there's an
* error, rbd_dev->spec->image_id will be filled in with
* a dynamically-allocated string, and rbd_dev->image_format
* will be set to either 1 or 2.
*/
ret = rbd_dev_image_id(rbd_dev);
if (ret)
@ -5029,16 +5032,18 @@ static ssize_t rbd_add(struct bus_type *bus,
rbd_dev->mapping.read_only = read_only;
rc = rbd_dev_device_setup(rbd_dev);
if (!rc)
return count;
if (rc) {
rbd_dev_image_release(rbd_dev);
goto err_out_module;
}
return count;
rbd_dev_image_release(rbd_dev);
err_out_rbd_dev:
rbd_dev_destroy(rbd_dev);
err_out_client:
rbd_put_client(rbdc);
err_out_args:
kfree(rbd_opts);
rbd_spec_put(spec);
err_out_module:
module_put(THIS_MODULE);