Instead of caching privilege information in 'kdbus_handle' (and thus
increasing the size of each handle by 8 byte), perform privilege checks
on the cached creds in file->f_cred inline. None of the functions that
need those checks is a fast path (in fact, it's just EP_MAKE that needs
it right now), so there's no need to cache the data.
If more functions need this check later on, we should turn it into a
small static helper.
Signed-off-by: David Herrmann <dh.herrm...@gmail.com>
---
ipc/kdbus/handle.c | 34 +++++++++++-----------------------
1 file changed, 11 insertions(+), 23 deletions(-)
diff --git a/ipc/kdbus/handle.c b/ipc/kdbus/handle.c
index a93c385..4d41ecf 100644
--- a/ipc/kdbus/handle.c
+++ b/ipc/kdbus/handle.c
@@ -264,7 +264,6 @@ enum kdbus_handle_type {
* @bus_owner: bus this handle owns
* @ep_owner: endpoint this handle owns
* @conn: connection this handle owns
- * @privileged: Flag to mark a handle as privileged
*/
struct kdbus_handle {
struct mutex lock;
@@ -275,8 +274,6 @@ struct kdbus_handle {
struct kdbus_ep *ep_owner;
struct kdbus_conn *conn;
};
-
- bool privileged:1;
};
static int kdbus_handle_open(struct inode *inode, struct file *file)
@@ -298,23 +295,6 @@ static int kdbus_handle_open(struct inode *inode, struct
file *file)
mutex_init(&handle->lock);
handle->type = KDBUS_HANDLE_NONE;
- if (node->type == KDBUS_NODE_ENDPOINT) {
- struct kdbus_ep *ep = kdbus_ep_from_node(node);
- struct kdbus_bus *bus = ep->bus;
-
- /*
- * A connection is privileged if it is opened on an endpoint
- * without custom policy and either:
- * * the user has CAP_IPC_OWNER in the domain user namespace
- * or
- * * the callers euid matches the uid of the bus creator
- */
- if (!ep->user &&
- (ns_capable(bus->domain->user_namespace, CAP_IPC_OWNER) ||
- uid_eq(file->f_cred->euid, bus->node.uid)))
- handle->privileged = true;
- }
-
file->private_data = handle;
ret = 0;
@@ -406,6 +386,7 @@ static long kdbus_handle_ioctl_ep(struct file *file,
unsigned int cmd,
struct kdbus_handle *handle = file->private_data;
struct kdbus_node *node = file_inode(file)->i_private;
struct kdbus_ep *ep, *file_ep = kdbus_ep_from_node(node);
+ struct kdbus_bus *bus = file_ep->bus;
struct kdbus_conn *conn;
int ret = 0;
@@ -413,14 +394,20 @@ static long kdbus_handle_ioctl_ep(struct file *file,
unsigned int cmd,
return -ESHUTDOWN;
switch (cmd) {
- case KDBUS_CMD_ENDPOINT_MAKE:
+ case KDBUS_CMD_ENDPOINT_MAKE: {
+ bool priv;
+
+ priv = uid_eq(file->f_cred->euid, bus->node.uid) ||
+ file_ns_capable(file, bus->domain->user_namespace,
+ CAP_IPC_OWNER);
+
/* creating custom endpoints is a privileged operation */
- if (!handle->privileged) {
+ if (file_ep->user || !priv) {
ret = -EPERM;
break;
}
- ep = kdbus_cmd_ep_make(file_ep->bus, buf);
+ ep = kdbus_cmd_ep_make(bus, buf);
if (IS_ERR_OR_NULL(ep)) {
ret = PTR_ERR_OR_ZERO(ep);
break;
@@ -429,6 +416,7 @@ static long kdbus_handle_ioctl_ep(struct file *file,
unsigned int cmd,
handle->ep_owner = ep;
ret = KDBUS_HANDLE_EP_OWNER;
break;
+ }
case KDBUS_CMD_HELLO:
conn = kdbus_cmd_hello(file_ep, file, buf);
--
2.5.0
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
