svcrdma: Fix race between svc_rdma_recvfrom thread and the dto_tasklet
RDMA_READ completions are kept on a separate queue from the general
I/O request queue. Since a separate lock is used to protect the RDMA_READ
completion queue, a race exists between the dto_tasklet and the
svc_rdma_recvfrom thread where the dto_tasklet sets the XPT_DATA
bit and adds I/O to the read-completion queue. Concurrently, the
recvfrom thread checks the generic queue, finds it empty and resets
the XPT_DATA bit. A subsequent svc_xprt_enqueue will fail to enqueue
the transport for I/O and cause the transport to "stall".
The fix is to protect both lists with the same lock and set the XPT_DATA
bit with this lock held.
Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
diff --git a/net/sunrpc/xprtrdma/svc_rdma_transport.c b/net/sunrpc/xprtrdma/svc_rdma_transport.c
index 19ddc38..900cb69 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_transport.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_transport.c
@@ -359,11 +359,11 @@
if (test_bit(RDMACTXT_F_LAST_CTXT, &ctxt->flags)) {
struct svc_rdma_op_ctxt *read_hdr = ctxt->read_hdr;
BUG_ON(!read_hdr);
+ spin_lock_bh(&xprt->sc_rq_dto_lock);
set_bit(XPT_DATA, &xprt->sc_xprt.xpt_flags);
- spin_lock_bh(&xprt->sc_read_complete_lock);
list_add_tail(&read_hdr->dto_q,
&xprt->sc_read_complete_q);
- spin_unlock_bh(&xprt->sc_read_complete_lock);
+ spin_unlock_bh(&xprt->sc_rq_dto_lock);
svc_xprt_enqueue(&xprt->sc_xprt);
}
svc_rdma_put_context(ctxt, 0);
@@ -428,7 +428,6 @@
init_waitqueue_head(&cma_xprt->sc_send_wait);
spin_lock_init(&cma_xprt->sc_lock);
- spin_lock_init(&cma_xprt->sc_read_complete_lock);
spin_lock_init(&cma_xprt->sc_rq_dto_lock);
cma_xprt->sc_ord = svcrdma_ord;