All of lore.kernel.org
 help / color / mirror / Atom feed
From: Chuck Lever <cel@kernel.org>
To: Anna Schumaker <anna@kernel.org>
Cc: Trond Myklebust <trond.myklebust@hammerspace.com>,
	<linux-nfs@vger.kernel.org>, <linux-rdma@vger.kernel.org>,
	Chuck Lever <chuck.lever@oracle.com>
Subject: [PATCH v2 7/8] xprtrdma: Scale receive batch size with credit window
Date: Thu,  5 Mar 2026 09:51:02 -0500	[thread overview]
Message-ID: <20260305145054.7096-17-cel@kernel.org> (raw)
In-Reply-To: <20260305145054.7096-10-cel@kernel.org>

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

The fixed RPCRDMA_MAX_RECV_BATCH of 7 results in frequent
small ib_post_recv batches during high-rate workloads. With
a 128-slot credit window, receives are reposted every 7th
completion, each batch incurring atomic serialization and a
doorbell write.

Replace the fixed batch constant with a per-endpoint value
scaled to 25% of the negotiated credit window. For a typical
128-credit connection this raises the batch from 7 to 32,
reducing doorbell frequency by roughly 4x and amortizing the
per-batch atomic and MMIO costs over a larger group of
receive WRs.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 net/sunrpc/xprtrdma/frwr_ops.c  | 3 ++-
 net/sunrpc/xprtrdma/verbs.c     | 2 +-
 net/sunrpc/xprtrdma/xprt_rdma.h | 1 +
 3 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c
index ef6a6ab9f940..ab7c46658c16 100644
--- a/net/sunrpc/xprtrdma/frwr_ops.c
+++ b/net/sunrpc/xprtrdma/frwr_ops.c
@@ -244,9 +244,10 @@ int frwr_query_device(struct rpcrdma_ep *ep, const struct ib_device *device)
 	}
 	ep->re_attr.cap.max_send_wr += RPCRDMA_BACKWARD_WRS;
 	ep->re_attr.cap.max_send_wr += 1; /* for ib_drain_sq */
+	ep->re_recv_batch = ep->re_max_requests >> 2;
 	ep->re_attr.cap.max_recv_wr = ep->re_max_requests;
 	ep->re_attr.cap.max_recv_wr += RPCRDMA_BACKWARD_WRS;
-	ep->re_attr.cap.max_recv_wr += RPCRDMA_MAX_RECV_BATCH;
+	ep->re_attr.cap.max_recv_wr += ep->re_recv_batch;
 	ep->re_attr.cap.max_recv_wr += 1; /* for ib_drain_rq */
 
 	ep->re_max_rdma_segs =
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index 90fd83f2d846..aecf9c0a153f 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -1374,7 +1374,7 @@ void rpcrdma_post_recvs(struct rpcrdma_xprt *r_xprt, int needed)
 	if (likely(ep->re_receive_count > needed))
 		goto out;
 	needed -= ep->re_receive_count;
-	needed += RPCRDMA_MAX_RECV_BATCH;
+	needed += ep->re_recv_batch;
 
 	if (atomic_inc_return(&ep->re_receiving) > 1)
 		goto out_dec;
diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h
index 37bba72065e8..f53a77472724 100644
--- a/net/sunrpc/xprtrdma/xprt_rdma.h
+++ b/net/sunrpc/xprtrdma/xprt_rdma.h
@@ -96,6 +96,7 @@ struct rpcrdma_ep {
 	struct rpcrdma_notification	re_rn;
 	int			re_receive_count;
 	unsigned int		re_max_requests; /* depends on device */
+	unsigned int		re_recv_batch;
 	unsigned int		re_inline_send;	/* negotiated */
 	unsigned int		re_inline_recv;	/* negotiated */
 
-- 
2.53.0


  parent reply	other threads:[~2026-03-05 14:51 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-03-05 14:50 [PATCH v2 0/8] Fix various races in xprtrdma Chuck Lever
2026-03-05 14:50 ` [PATCH v2 1/8] xprtrdma: Decrement re_receiving on the early exit paths Chuck Lever
2026-03-05 14:52   ` Chuck Lever
2026-03-05 14:50 ` [PATCH v2 2/8] xprtrdma: Close sendctx get/put race that can block a transport Chuck Lever
2026-03-05 14:50 ` [PATCH v2 3/8] xprtrdma: Avoid 250 ms delay on backlog wakeup Chuck Lever
2026-03-05 14:50 ` [PATCH v2 4/8] xprtrdma: Close lost-wakeup race in xprt_rdma_alloc_slot Chuck Lever
2026-03-05 14:51 ` [PATCH v2 5/8] xprtrdma: Decouple frwr_wp_create from frwr_map Chuck Lever
2026-03-05 14:51 ` [PATCH v2 6/8] xprtrdma: Replace rpcrdma_mr_seg with xdr_buf cursor Chuck Lever
2026-03-06 16:28   ` Anna Schumaker
2026-03-06 16:30     ` Chuck Lever
2026-03-05 14:51 ` Chuck Lever [this message]
2026-03-05 14:51 ` [PATCH v2 8/8] xprtrdma: Post receive buffers after RPC completion Chuck Lever

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=20260305145054.7096-17-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 \
    --cc=trond.myklebust@hammerspace.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.