CVE-2026-72473 PUBLISHED

xprtrdma: Decouple req recycling from RPC completion

Assigner: Linux
Reserved: 09.08.2026 Published: 15.08.2026 Updated: 17.08.2026

In the Linux kernel, the following vulnerability has been resolved:

xprtrdma: Decouple req recycling from RPC completion

rl_kref formerly served two distinct lifetimes through a single refcount: it gated when a Reply could wake its RPC task, and it gated when an rpcrdma_req could return to its free pool. The marshal path took the Send-side reference only when SGEs needed DMA-unmap (sc_unmap_count > 0), which made a Send carrying only pre-registered buffers an exception: the Reply handler dropped rl_kref from 1 to 0 and freed the req while the HCA might still be DMA-reading from its send buffer.

Give rl_kref a narrower job. The RPC layer takes one reference when slot allocation hands a req out. rpcrdma_prepare_send_sges() takes a Send-side reference unconditionally after WR preparation succeeds. xprt_rdma_free_slot() and xprt_rdma_bc_free_rqst() drop the RPC-layer reference; rpcrdma_sendctx_unmap() drops the Send-side reference. The req returns to its free pool only after both owners have signed off.

The existing kref_init(&req->rl_kref) call in rpcrdma_prepare_send_sges() is removed. Initialization moves to the slot-allocation paths (xprt_rdma_alloc_slot and rpcrdma_bc_rqst_get), and the release callback re-arms rl_kref before the req returns to a free pool. A re-init in the marshal path would discard the RPC-layer reference that already exists on entry.

Three invariants follow:

  • Any rpcrdma_req held by an rpc_rqst has rl_kref >= 1. xprt_rdma_alloc_slot(), rpcrdma_bc_rqst_get(), and the backlog-wake branch in xprt_rdma_alloc_slot() each kref_init rl_kref before publishing the req. Without this invariant, an RPC task that aborts between slot allocation and marshal (gss_refresh failure or signal during call_connect, for example) would drive xprt_release() -> xprt_rdma_free_slot() -> kref_put against a refcount of zero, saturating refcount_t and stranding the slot.

  • The Send-side reference is taken only after WR prep succeeds. A mapping failure in rpcrdma_prepare_send_sges() runs rpcrdma_sendctx_cancel(), which DMA-unmaps the sendctx and clears sc_req without touching rl_kref. The sendctx ring walks in rpcrdma_sendctx_put_locked() and rpcrdma_sendctxs_destroy() skip entries with sc_req == NULL, so a burst of -EIO marshal failures cannot hold reqs off rb_send_bufs.

  • The release callback re-arms rl_kref so the next consumer enters with the invariant satisfied.

Replies now complete the RPC directly. rpcrdma_reply_handler() calls rpcrdma_complete_rqst() in place of kref_put on the non-LocalInv branch. The LocalInv branch already completes the RPC from frwr_unmap_async() and is unaffected.

Because Send-side references can now outlive RPC completion, connection teardown drains sendctx entries whose unsignaled Sends never had a later signaled completion to walk the ring. rpcrdma_sendctxs_destroy() walks the active range and runs rpcrdma_sendctx_unmap() on each entry with a non-NULL sc_req before the request buffers are reset, and is moved ahead of rpcrdma_reqs_reset() in rpcrdma_xprt_disconnect() so the reqs are still in their pre-reset state when the Send-side refs are released.

The drain creates a teardown-ordering hazard on the backchannel path. With the new lifetime, releasing a bc_prealloc req from rpcrdma_req_release() re-adds it to bc_pa_list. The disconnect in xprt_rdma_destroy() runs after xprt_destroy_backchannel() has already emptied bc_pa_list, so the drained reqs would otherwise leak. xprt_rdma_destroy() now runs xprt_rdma_bc_destroy(xprt, 0) a second time after the disconnect to reclaim them.

Metrics

CVSS Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
CVSS Score: 9.8

AV:N - The bug is in the RPC-over-RDMA client (xprtrdma); a remote NFS/RDMA peer delivers Receive completions that invoke rpcrdma_reply_handler() over InfiniBand/RoCE/iWARP, so exploitation is via network protocol traffic from a malicious or compromised server. AC:L - For inline Sends using only pre-registered buffers (sc_unmap_count==0), the reply path completes the RPC and returns the req to the free pool before Send completion; a remote peer can reliably trigger this by replying immediately to normal small NFS RPCs. PR:N - Exploitation requires no privileges on the victim host; a remote malicious or compromised NFS/RDMA server can send crafted/fast replies over the established RDMA connection to drive the vulnerable completion path. UI:N - No end-user interaction is needed at exploit time; once a host uses NFS-over-RDMA, triggering the bug is automatic during ordinary client I/O against the attacker-controlled server. S:U - The flaw causes kernel heap UAF/DMA corruption within the NFS client's kernel context and does not by itself cross hypervisor, VM, or sandbox security boundaries. C:H - Recycling rpcrdma_req while the HCA may still DMA-read its send buffers is a use-after-free that can expose or leak kernel memory when the slot is reallocated and buffers are reused under active DMA. I:H - Premature return of rpcrdma_req to rb_send_bufs while Send WRs are in flight permits heap corruption and typical UAF exploitation paths for arbitrary kernel memory writes or control-flow hijack. A:H - Concurrent DMA against freed/reused request buffers can provoke kernel oops/panic or wedged RPC/RDMA transport state, giving high availability impact on NFS-over-RDMA clients.

Product Status

Vendor Linux
Product Linux
Versions Default: unaffected
  • affected from 0ab115237025f5e379620bbcd56a02697d07b002 to 740975054a1970c0cf15f70ac39724a064f45847 (excl.)
  • affected from 0ab115237025f5e379620bbcd56a02697d07b002 to 9f3d9b68c1c6c51746e5ecdb52b2e6a2901de37e (excl.)
  • affected from 0ab115237025f5e379620bbcd56a02697d07b002 to e7632089523acddcdd8f090ad19e96fb3107b04d (excl.)
  • affected from 0ab115237025f5e379620bbcd56a02697d07b002 to 53442c7d0c888e51b8bc3da196970a669cc6b294 (excl.)
  • affected from 0ab115237025f5e379620bbcd56a02697d07b002 to 8203f760a72bd39a3b66bc4eff0aa272a99fe22b (excl.)
  • affected from 0ab115237025f5e379620bbcd56a02697d07b002 to e786233d2e0bbff9a82e43f02ae3a46ab4b08ec3 (excl.)
Vendor Linux
Product Linux
Versions Default: affected
  • Version 5.3 is affected
  • unaffected from 0 to 5.3 (excl.)
  • unaffected from 6.1.178 to 6.1.* (incl.)
  • unaffected from 6.6.145 to 6.6.* (incl.)
  • unaffected from 6.12.97 to 6.12.* (incl.)
  • unaffected from 6.18.40 to 6.18.* (incl.)
  • unaffected from 7.1.5 to 7.1.* (incl.)
  • unaffected from 7.2 to * (incl.)

References