svcrdma: support Remote Invalidation

Support Remote Invalidation. A private message is exchanged with
the client upon RDMA transport connect that indicates whether
Send With Invalidation may be used by the server to send RPC
replies. The invalidate_rkey is arbitrarily chosen from among
rkeys present in the RPC-over-RDMA header's chunk lists.

Send With Invalidate improves performance only when clients can
recognize, while processing an RPC reply, that an rkey has already
been invalidated. That has been submitted as a separate change.

In the future, the RPC-over-RDMA protocol might support Remote
Invalidation properly. The protocol needs to enable signaling
between peers to indicate when Remote Invalidation can be used
for each individual RPC.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>

2 files changed, 64 insertions, 6 deletions

diff --git a/net/sunrpc/xprtrdma/svc_rdma_sendto.c b/net/sunrpc/xprtrdma/svc_rdma_sendto.c
index 3b95b19..f5a91ed 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_sendto.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_sendto.c
@@ -225,6 +225,48 @@ svc_rdma_get_reply_array(struct rpcrdma_msg *rmsgp,
 	return rp_ary;
 }
 
+/* RPC-over-RDMA Version One private extension: Remote Invalidation.
+ * Responder's choice: requester signals it can handle Send With
+ * Invalidate, and responder chooses one rkey to invalidate.
+ *
+ * Find a candidate rkey to invalidate when sending a reply.  Picks the
+ * first rkey it finds in the chunks lists.
+ *
+ * Returns zero if RPC's chunk lists are empty.
+ */
+static u32 svc_rdma_get_inv_rkey(struct rpcrdma_msg *rdma_argp,
+				 struct rpcrdma_write_array *wr_ary,
+				 struct rpcrdma_write_array *rp_ary)
+{
+	struct rpcrdma_read_chunk *rd_ary;
+	struct rpcrdma_segment *arg_ch;
+	u32 inv_rkey;
+
+	inv_rkey = 0;
+
+	rd_ary = svc_rdma_get_read_chunk(rdma_argp);
+	if (rd_ary) {
+		inv_rkey = be32_to_cpu(rd_ary->rc_target.rs_handle);
+		goto out;
+	}
+
+	if (wr_ary && be32_to_cpu(wr_ary->wc_nchunks)) {
+		arg_ch = &wr_ary->wc_array[0].wc_target;
+		inv_rkey = be32_to_cpu(arg_ch->rs_handle);
+		goto out;
+	}
+
+	if (rp_ary && be32_to_cpu(rp_ary->wc_nchunks)) {
+		arg_ch = &rp_ary->wc_array[0].wc_target;
+		inv_rkey = be32_to_cpu(arg_ch->rs_handle);
+		goto out;
+	}
+
+out:
+	dprintk("svcrdma: Send With Invalidate rkey=%08x\n", inv_rkey);
+	return inv_rkey;
+}
+
 /* Assumptions:
  * - The specified write_len can be represented in sc_max_sge * PAGE_SIZE
  */
@@ -464,7 +506,8 @@ static int send_reply(struct svcxprt_rdma *rdma,
 		      struct page *page,
 		      struct rpcrdma_msg *rdma_resp,
 		      struct svc_rdma_req_map *vec,
-		      int byte_count)
+		      int byte_count,
+		      u32 inv_rkey)
 {
 	struct svc_rdma_op_ctxt *ctxt;
 	struct ib_send_wr send_wr;
@@ -535,7 +578,11 @@ static int send_reply(struct svcxprt_rdma *rdma,
 	send_wr.wr_cqe = &ctxt->cqe;
 	send_wr.sg_list = ctxt->sge;
 	send_wr.num_sge = sge_no;
-	send_wr.opcode = IB_WR_SEND;
+	if (inv_rkey) {
+		send_wr.opcode = IB_WR_SEND_WITH_INV;
+		send_wr.ex.invalidate_rkey = inv_rkey;
+	} else
+		send_wr.opcode = IB_WR_SEND;
 	send_wr.send_flags =  IB_SEND_SIGNALED;
 
 	ret = svc_rdma_send(rdma, &send_wr);
@@ -567,6 +614,7 @@ int svc_rdma_sendto(struct svc_rqst *rqstp)
 	int inline_bytes;
 	struct page *res_page;
 	struct svc_rdma_req_map *vec;
+	u32 inv_rkey;
 
 	dprintk("svcrdma: sending response for rqstp=%p\n", rqstp);
 
@@ -577,6 +625,10 @@ int svc_rdma_sendto(struct svc_rqst *rqstp)
 	wr_ary = svc_rdma_get_write_array(rdma_argp);
 	rp_ary = svc_rdma_get_reply_array(rdma_argp, wr_ary);
 
+	inv_rkey = 0;
+	if (rdma->sc_snd_w_inv)
+		inv_rkey = svc_rdma_get_inv_rkey(rdma_argp, wr_ary, rp_ary);
+
 	/* Build an req vec for the XDR */
 	vec = svc_rdma_get_req_map(rdma);
 	ret = svc_rdma_map_xdr(rdma, &rqstp->rq_res, vec, wr_ary != NULL);
@@ -619,7 +671,7 @@ int svc_rdma_sendto(struct svc_rqst *rqstp)
 		goto err1;
 
 	ret = send_reply(rdma, rqstp, res_page, rdma_resp, vec,
-			 inline_bytes);
+			 inline_bytes, inv_rkey);
 	if (ret < 0)
 		goto err0;
 
diff --git a/net/sunrpc/xprtrdma/svc_rdma_transport.c b/net/sunrpc/xprtrdma/svc_rdma_transport.c
index f51e98a..b2464fc 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_transport.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_transport.c
@@ -657,9 +657,14 @@ svc_rdma_parse_connect_private(struct svcxprt_rdma *newxprt,
 	if (pmsg &&
 	    pmsg->cp_magic == rpcrdma_cmp_magic &&
 	    pmsg->cp_version == RPCRDMA_CMP_VERSION) {
-		dprintk("svcrdma: client send_size %u, recv_size %u\n",
+		newxprt->sc_snd_w_inv = pmsg->cp_flags &
+					RPCRDMA_CMP_F_SND_W_INV_OK;
+
+		dprintk("svcrdma: client send_size %u, recv_size %u "
+			"remote inv %ssupported\n",
 			rpcrdma_decode_buffer_size(pmsg->cp_send_size),
-			rpcrdma_decode_buffer_size(pmsg->cp_recv_size));
+			rpcrdma_decode_buffer_size(pmsg->cp_recv_size),
+			newxprt->sc_snd_w_inv ? "" : "un");
 	}
 }
 
@@ -1093,7 +1098,8 @@ static struct svc_xprt *svc_rdma_accept(struct svc_xprt *xprt)
 			dev->attrs.max_fast_reg_page_list_len;
 		newxprt->sc_dev_caps |= SVCRDMA_DEVCAP_FAST_REG;
 		newxprt->sc_reader = rdma_read_chunk_frmr;
-	}
+	} else
+		newxprt->sc_snd_w_inv = false;
 
 	/*
 	 * Determine if a DMA MR is required and if so, what privs are required