Linux RDMA and InfiniBand development
 help / color / mirror / Atom feed
From: Chuck Lever <cel@kernel.org>
To: Anna Schumaker <anna@kernel.org>
Cc: <linux-nfs@vger.kernel.org>, <linux-rdma@vger.kernel.org>,
	Chuck Lever <chuck.lever@oracle.com>
Subject: [PATCH v2 3/3] xprtrdma: Add request-pool slack for delayed recycling
Date: Fri, 22 May 2026 20:02:52 -0400	[thread overview]
Message-ID: <20260523000252.465074-4-cel@kernel.org> (raw)
In-Reply-To: <20260523000252.465074-1-cel@kernel.org>

From: Chuck Lever <chuck.lever@oracle.com>

After the previous patch gates req recycling on Send completion,
a completed RPC's rpcrdma_req can remain pinned by the sendctx
ring until the next signaled Send completion releases it. The
transmitted-RPC ceiling is unchanged: xprt_request_get_cong()
gates Sends against xprt->cwnd, the RPC/RDMA credit window fed
by server-granted credits and capped at re_max_requests. The
req pool, however, must exceed max_reqs by enough that this
recycle delay does not stall a slot allocation that the credit
window would admit.

The headroom is bounded. rpcrdma_ep_create() sets re_send_batch
to re_max_requests >> 3 -- one in every eight Sends is signaled
-- so at most re_send_batch unsignaled Sends can be outstanding
before the next signaled completion releases them. That equals
max_reqs / 8 reqs in the worst case.

Allocate max_reqs + DIV_ROUND_UP(max_reqs, 8) request objects
and name the slack calculation at the allocation site so the
1/8 bound stays tied to the Send-signaling batch size.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 net/sunrpc/xprtrdma/verbs.c | 17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)

diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index 97b8b2376602..a83e51f7f15e 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -1080,6 +1080,18 @@ static void rpcrdma_reps_destroy(struct rpcrdma_buffer *buf)
 	spin_unlock(&buf->rb_lock);
 }
 
+static unsigned int rpcrdma_req_pool_slack(unsigned int max_reqs)
+{
+	/* The sendctx ring can hold up to one Send-signaling batch
+	 * (re_send_batch, set by rpcrdma_ep_create() to re_max_requests >> 3)
+	 * of unfinished Sends. Each pins its req until a signaled Send
+	 * completion releases the sendctx. Size the pool above max_reqs
+	 * by that batch so the recycle delay does not stall a slot
+	 * allocation that the RPC/RDMA credit window would admit.
+	 */
+	return DIV_ROUND_UP(max_reqs, 8);
+}
+
 /**
  * rpcrdma_buffer_create - Create initial set of req/rep objects
  * @r_xprt: transport instance to (re)initialize
@@ -1089,6 +1101,7 @@ static void rpcrdma_reps_destroy(struct rpcrdma_buffer *buf)
 int rpcrdma_buffer_create(struct rpcrdma_xprt *r_xprt)
 {
 	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
+	unsigned int max_reqs;
 	int i, rc;
 
 	buf->rb_bc_srv_max_requests = 0;
@@ -1102,7 +1115,9 @@ int rpcrdma_buffer_create(struct rpcrdma_xprt *r_xprt)
 	INIT_LIST_HEAD(&buf->rb_all_reps);
 
 	rc = -ENOMEM;
-	for (i = 0; i < r_xprt->rx_xprt.max_reqs; i++) {
+	max_reqs = r_xprt->rx_xprt.max_reqs;
+	max_reqs += rpcrdma_req_pool_slack(max_reqs);
+	for (i = 0; i < max_reqs; i++) {
 		struct rpcrdma_req *req;
 
 		req = rpcrdma_req_create(r_xprt,
-- 
2.54.0


      parent reply	other threads:[~2026-05-23  0:02 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-05-23  0:02 [PATCH v2 0/3] xprtrdma: Decouple req recycling from RPC completion Chuck Lever
2026-05-23  0:02 ` [PATCH v2 1/3] xprtrdma: Use sendctx DMA state for Send signaling Chuck Lever
2026-05-23  0:02 ` [PATCH v2 2/3] xprtrdma: Decouple req recycling from RPC completion Chuck Lever
2026-05-23  0:02 ` Chuck Lever [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260523000252.465074-4-cel@kernel.org \
    --to=cel@kernel.org \
    --cc=anna@kernel.org \
    --cc=chuck.lever@oracle.com \
    --cc=linux-nfs@vger.kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox