Linux-NFS Archive on lore.kernel.org
 help / color / Atom feed
From: Chuck Lever <chuck.lever@oracle.com>
To: linux-rdma@vger.kernel.org, linux-nfs@vger.kernel.org
Subject: [PATCH v1 03/18] xprtrdma: Boost maximum transport header size
Date: Tue, 06 Aug 2019 11:54:02 -0400
Message-ID: <20190806155402.9529.80840.stgit@manet.1015granger.net> (raw)
In-Reply-To: <20190806155246.9529.14571.stgit@manet.1015granger.net>

Although I haven't seen any performance results that justify it,
I've received several complaints that NFS/RDMA no longer supports
a maximum rsize and wsize of 1MB. These days it is somewhat smaller.

To simplify the logic that determines whether a chunk list is
necessary, the implementation uses a fixed maximum size of the
transport header. Currently that maximum size is 256 bytes, one
quarter of the default inline threshold size for RPC/RDMA v1.

Since commit a78868497c2e ("xprtrdma: Reduce max_frwr_depth"), the
size of chunks is also smaller to take advantage of inline page
lists in MR data structures.

The combination of these two design choices has reduced the maximum
NFS rsize and wsize that can be used for most RNIC/HCAs. Increasing
the maximum transport header size and the maximum number of RDMA
segments it can contain increases the negotiated maximum rsize/wsize
on common RNIC/HCAs.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 net/sunrpc/xprtrdma/xprt_rdma.h |   17 ++++++++---------
 1 file changed, 8 insertions(+), 9 deletions(-)

diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h
index 92ce09f..f9071fb 100644
--- a/net/sunrpc/xprtrdma/xprt_rdma.h
+++ b/net/sunrpc/xprtrdma/xprt_rdma.h
@@ -158,25 +158,24 @@ static inline void *rdmab_data(const struct rpcrdma_regbuf *rb)
 
 /* To ensure a transport can always make forward progress,
  * the number of RDMA segments allowed in header chunk lists
- * is capped at 8. This prevents less-capable devices and
+ * is capped at 12. This prevents less-capable devices and
  * memory registrations from overrunning the Send buffer
  * while building chunk lists.
  *
  * Elements of the Read list take up more room than the
- * Write list or Reply chunk. 8 read segments means the Read
- * list (or Write list or Reply chunk) cannot consume more
- * than
+ * Write list or Reply chunk. 12 read segments means the
+ * chunk lists cannot consume more than
  *
- * ((8 + 2) * read segment size) + 1 XDR words, or 244 bytes.
+ * ((12 + 2) * read segment size) + 1 XDR words, or 340 bytes.
  *
- * And the fixed part of the header is another 24 bytes.
+ * The fixed part of the header is another 24 bytes.
  *
  * The smallest inline threshold is 1024 bytes, ensuring that
- * at least 750 bytes are available for RPC messages.
+ * at least 650 bytes are available for RPC message bodies.
  */
 enum {
-	RPCRDMA_MAX_HDR_SEGS = 8,
-	RPCRDMA_HDRBUF_SIZE = 256,
+	RPCRDMA_MAX_HDR_SEGS = 12,
+	RPCRDMA_HDRBUF_SIZE = 512,
 };
 
 /*


  parent reply index

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-06 15:53 [PATCH v1 00/18] NFS/RDMA patches Chuck Lever
2019-08-06 15:53 ` [PATCH v1 01/18] xprtrdma: Refresh the documenting comment in frwr_ops.c Chuck Lever
2019-08-06 15:53 ` [PATCH v1 02/18] xprtrdma: Fix calculation of ri_max_segs again Chuck Lever
2019-08-06 15:54 ` Chuck Lever [this message]
2019-08-06 15:54 ` [PATCH v1 04/18] xprtrdma: Boost client's max slot table size to match Linux server Chuck Lever
2019-08-06 15:54 ` [PATCH v1 05/18] xprtrdma: Rename CQE field in Receive trace points Chuck Lever
2019-08-06 15:54 ` [PATCH v1 06/18] xprtrdma: Rename rpcrdma_buffer::rb_all Chuck Lever
2019-08-06 15:54 ` [PATCH v1 07/18] xprtrdma: Toggle XPRT_CONGESTED in xprtrdma's slot methods Chuck Lever
2019-08-06 15:54 ` [PATCH v1 08/18] xprtrdma: Simplify rpcrdma_mr_pop Chuck Lever
2019-08-06 15:54 ` [PATCH v1 09/18] xprtrdma: Combine rpcrdma_mr_put and rpcrdma_mr_unmap_and_put Chuck Lever
2019-08-06 15:54 ` [PATCH v1 10/18] xprtrdma: Move rpcrdma_mr_get out of frwr_map Chuck Lever
2019-08-06 15:54 ` [PATCH v1 11/18] xprtrdma: Ensure creating an MR does not trigger FS writeback Chuck Lever
2019-08-06 15:54 ` [PATCH v1 12/18] xprtrdma: Cache free MRs in each rpcrdma_req Chuck Lever
2019-08-06 15:54 ` [PATCH v1 13/18] xprtrdma: Remove rpcrdma_buffer::rb_mrlock Chuck Lever
2019-08-06 15:55 ` [PATCH v1 14/18] xprtrdma: Use an llist to manage free rpcrdma_reps Chuck Lever
2019-08-06 15:55 ` [PATCH v1 15/18] xprtrdma: Clean up xprt_rdma_set_connect_timeout() Chuck Lever
2019-08-06 15:55 ` [PATCH v1 16/18] xprtdma: Fix bc_max_slots return value Chuck Lever
2019-08-06 15:55 ` [PATCH v1 17/18] xprtrdma: Inline XDR chunk encoder functions Chuck Lever
2019-08-06 15:55 ` [PATCH v1 18/18] xprtrdma: Optimize rpcrdma_post_recvs() Chuck Lever

Reply instructions:

You may reply publically 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=20190806155402.9529.80840.stgit@manet.1015granger.net \
    --to=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

Linux-NFS Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-nfs/0 linux-nfs/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-nfs linux-nfs/ https://lore.kernel.org/linux-nfs \
		linux-nfs@vger.kernel.org
	public-inbox-index linux-nfs

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-nfs


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git