Linux kernel use-after-free in RDS RDMA route failure handling
Linux
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
|
Linux
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
|
≥ 55b7ed0b582f9ff4bf3b9b44969cddda8c6763ea · < 91ece3efa064f0b3b3c67b9b045d9eeb4a034e43| No fix yet |
| Linux Linux |
|
|
In the Linux kernel, the following vulnerability has been resolved:
net/rds: don't let the rdma_cm destroy an id RDS still owns on route failure
rds_rdma_cm_event_handler_cmn() hands the return value of rdma_resolve_route() straight back to the rdma_cm on RDMA_CM_EVENT_ADDR_RESOLVED. A synchronous failure there - -ENOMEM from the route work allocation, an SA query that cannot be set up, a RoCE route with no usable device - makes addr_handler() destroy the id the event was delivered on. That id is ic->i_cm_id, and nothing clears the pointer: the connection sits in RDS_CONN_CONNECTING with a freed id until its shutdown calls rdma_disconnect() and rdma_destroy_id() on it.
The ROUTE_RESOLVED case already avoids this: rds_ib_cm_initiate_connect() forces a zero return while ic->i_cm_id == cm_id, as the comment there explains. Do the same here - drop the connection and return 0 - so the id stays RDS's to destroy from the shutdown, and the reconnect gets a fresh one.
From the CVE record · Linux
The Linux kernel’s RDS RDMA connection handler can retain a pointer to a freed connection ID when route resolution fails synchronously. Later shutdown can use that freed ID. The input does not specify attacker access requirements or whether user interaction is needed.
Engineers running Linux kernels with RDS RDMA support should check their versions; affected versions were not provided.