Linux NFS development
 help / color / mirror / Atom feed
From: Tom Talpey <talpey@netapp.com>
To: linux-nfs@vger.kernel.org
Subject: [PATCH 05/15] RPC/RDMA: fix connection IRD/ORD setting
Date: Wed, 08 Oct 2008 11:47:44 -0400	[thread overview]
Message-ID: <20081008154744.1336.20909.stgit@tmt3.nane.netapp.com> (raw)
In-Reply-To: <20081008154506.1336.59892.stgit-pfX4bTJKMULWwzOYslWYilaTQe2KTcn/@public.gmane.org>

From: Tom Tucker <talpey@netapp.com>

This logic sets the connection parameter that configures the local device
and informs the remote peer how many concurrent incoming RDMA_READ
requests are supported. The original logic didn't really do what was
intended for two reasons:

- The max number supported by the device is typically smaller than
any one factor in the calculation used, and

- The field in the connection parameter structure where the value is
stored is a u8 and always overflows for the default settings.

So what really happens is the value requested for responder resources
is the left over 8 bits from the "desired value". If the desired value
happened to be a multiple of 256, the result was zero and it wouldn't
connect at all.

Given the above and the fact that max_requests is almost always larger
than the max responder resources supported by the adapter, this patch
simplifies this logic and simply requests the max supported by the device,
subject to a reasonable limit.

This bug was found by Jim Schutt at Sandia.

Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
Signed-off-by: Tom Talpey <talpey@netapp.com>
---

 net/sunrpc/xprtrdma/verbs.c |   51 ++++++++++++-------------------------------
 1 files changed, 14 insertions(+), 37 deletions(-)

diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index 39a1652..e3fe905 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -705,30 +705,13 @@ rpcrdma_ep_create(struct rpcrdma_ep *ep, struct rpcrdma_ia *ia,
 	ep->rep_remote_cma.private_data_len = 0;
 
 	/* Client offers RDMA Read but does not initiate */
-	switch (ia->ri_memreg_strategy) {
-	case RPCRDMA_BOUNCEBUFFERS:
+	ep->rep_remote_cma.initiator_depth = 0;
+	if (ia->ri_memreg_strategy == RPCRDMA_BOUNCEBUFFERS)
 		ep->rep_remote_cma.responder_resources = 0;
-		break;
-	case RPCRDMA_MTHCAFMR:
-	case RPCRDMA_REGISTER:
-	case RPCRDMA_FRMR:
-		ep->rep_remote_cma.responder_resources = cdata->max_requests *
-				(RPCRDMA_MAX_DATA_SEGS / 8);
-		break;
-	case RPCRDMA_MEMWINDOWS:
-	case RPCRDMA_MEMWINDOWS_ASYNC:
-#if RPCRDMA_PERSISTENT_REGISTRATION
-	case RPCRDMA_ALLPHYSICAL:
-#endif
-		ep->rep_remote_cma.responder_resources = cdata->max_requests *
-				(RPCRDMA_MAX_DATA_SEGS / 2);
-		break;
-	default:
-		break;
-	}
-	if (ep->rep_remote_cma.responder_resources > devattr.max_qp_rd_atom)
+	else if (devattr.max_qp_rd_atom > 32)	/* arbitrary but <= 255 */
+		ep->rep_remote_cma.responder_resources = 32;
+	else
 		ep->rep_remote_cma.responder_resources = devattr.max_qp_rd_atom;
-	ep->rep_remote_cma.initiator_depth = 0;
 
 	ep->rep_remote_cma.retry_count = 7;
 	ep->rep_remote_cma.flow_control = 0;
@@ -858,14 +841,6 @@ if (strnicmp(ia->ri_id->device->dma_device->bus->name, "pci", 3) == 0) {
 	}
 }
 
-	/* Theoretically a client initiator_depth > 0 is not needed,
-	 * but many peers fail to complete the connection unless they
-	 * == responder_resources! */
-	if (ep->rep_remote_cma.initiator_depth !=
-				ep->rep_remote_cma.responder_resources)
-		ep->rep_remote_cma.initiator_depth =
-			ep->rep_remote_cma.responder_resources;
-
 	ep->rep_connected = 0;
 
 	rc = rdma_connect(ia->ri_id, &ep->rep_remote_cma);
@@ -894,14 +869,16 @@ if (strnicmp(ia->ri_id->device->dma_device->bus->name, "pci", 3) == 0) {
 	if (ep->rep_connected <= 0) {
 		/* Sometimes, the only way to reliably connect to remote
 		 * CMs is to use same nonzero values for ORD and IRD. */
-		ep->rep_remote_cma.initiator_depth =
-					ep->rep_remote_cma.responder_resources;
-		if (ep->rep_remote_cma.initiator_depth == 0)
-			++ep->rep_remote_cma.initiator_depth;
-		if (ep->rep_remote_cma.responder_resources == 0)
-			++ep->rep_remote_cma.responder_resources;
-		if (retry_count++ == 0)
+		if (retry_count++ <= RDMA_CONNECT_RETRY_MAX + 1 &&
+		    (ep->rep_remote_cma.responder_resources == 0 ||
+		     ep->rep_remote_cma.initiator_depth !=
+				ep->rep_remote_cma.responder_resources)) {
+			if (ep->rep_remote_cma.responder_resources == 0)
+				ep->rep_remote_cma.responder_resources = 1;
+			ep->rep_remote_cma.initiator_depth =
+				ep->rep_remote_cma.responder_resources;
 			goto retry;
+		}
 		rc = ep->rep_connected;
 	} else {
 		dprintk("RPC:       %s: connected\n", __func__);


  parent reply	other threads:[~2008-10-08 16:22 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-10-08 15:46 [PATCH 00/15] RPC/RDMA patchset for next merge window Tom Talpey
     [not found] ` <20081008154506.1336.59892.stgit-pfX4bTJKMULWwzOYslWYilaTQe2KTcn/@public.gmane.org>
2008-10-08 15:47   ` [PATCH 01/15] RPC/RDMA: refactor the inline memory registration code Tom Talpey
2008-10-08 15:47   ` [PATCH 02/15] RPC/RDMA: add data types and new FRMR memory registration enum Tom Talpey
     [not found]     ` <20081008154713.1336.41538.stgit-pfX4bTJKMULWwzOYslWYilaTQe2KTcn/@public.gmane.org>
2008-10-08 17:23       ` Trond Myklebust
2008-10-08 17:30         ` Talpey, Thomas
     [not found]           ` <RTPCLUEXC1-PRDmcarc00000072-rtwIt2gI0FxT+ZUat5FNkAK/GNPrWCqfQQ4Iyu8u01E@public.gmane.org>
2008-10-08 17:40             ` Trond Myklebust
2008-10-08 17:55             ` J. Bruce Fields
2008-10-08 17:58               ` Talpey, Thomas
2008-10-08 15:47   ` [PATCH 03/15] RPC/RDMA: check selected memory registration mode at runtime Tom Talpey
     [not found]     ` <20081008154723.1336.57976.stgit-pfX4bTJKMULWwzOYslWYilaTQe2KTcn/@public.gmane.org>
2008-10-08 17:22       ` Trond Myklebust
2008-10-08 17:29         ` Talpey, Thomas
     [not found]           ` <RTPCLUEXC1-PRD8yfog00000071-rtwIt2gI0FxT+ZUat5FNkAK/GNPrWCqfQQ4Iyu8u01E@public.gmane.org>
2008-10-08 17:40             ` Trond Myklebust
2008-10-08 15:47   ` [PATCH 04/15] RPC/RDMA: support FRMR client memory registration Tom Talpey
2008-10-08 15:47   ` Tom Talpey [this message]
     [not found]     ` <20081008154744.1336.20909.stgit-pfX4bTJKMULWwzOYslWYilaTQe2KTcn/@public.gmane.org>
2008-10-08 17:26       ` [PATCH 05/15] RPC/RDMA: fix connection IRD/ORD setting Trond Myklebust
2008-10-08 17:32         ` Talpey, Thomas
2008-10-08 15:47   ` [PATCH 06/15] RPC/RDMA: suppress retransmit on RPC/RDMA clients Tom Talpey
2008-10-08 15:48   ` [PATCH 07/15] RPC/RDMA: maintain the RPC task bytes-sent statistic Tom Talpey
2008-10-08 15:48   ` [PATCH 08/15] RPC/RDMA: avoid an oops due to disconnect racing with async upcalls Tom Talpey
2008-10-08 15:48   ` [PATCH 09/15] RPC/RDMA: adhere to protocol for unpadded client trailing write chunks Tom Talpey
     [not found]     ` <20081008154825.1336.79549.stgit-pfX4bTJKMULWwzOYslWYilaTQe2KTcn/@public.gmane.org>
2008-10-08 17:29       ` Trond Myklebust
2008-10-08 17:33         ` Talpey, Thomas
2008-10-08 15:48   ` [PATCH 10/15] RPC/RDMA: return a consistent error to mount, when connect fails Tom Talpey
     [not found]     ` <20081008154835.1336.85484.stgit-pfX4bTJKMULWwzOYslWYilaTQe2KTcn/@public.gmane.org>
2008-10-08 17:31       ` Trond Myklebust
2008-10-08 17:40         ` Talpey, Thomas
     [not found]           ` <RTPCLUEXC1-PRDbpH7100000075-rtwIt2gI0FxT+ZUat5FNkAK/GNPrWCqfQQ4Iyu8u01E@public.gmane.org>
2008-10-08 17:43             ` Trond Myklebust
2008-10-08 19:56               ` Talpey, Thomas
2008-10-08 15:48   ` [PATCH 11/15] RPC/RDMA: fix connect/reconnect resource leak Tom Talpey
2008-10-08 15:48   ` [PATCH 12/15] RPC/RDMA: correct a 5 second pause on reconnecting to an idle server Tom Talpey
     [not found]     ` <20081008154856.1336.18339.stgit-pfX4bTJKMULWwzOYslWYilaTQe2KTcn/@public.gmane.org>
2008-10-08 17:35       ` Trond Myklebust
2008-10-08 17:51         ` Talpey, Thomas
     [not found]           ` <RTPCLUEXC1-PRDjbDt300000076-rtwIt2gI0FxT+ZUat5FNkAK/GNPrWCqfQQ4Iyu8u01E@public.gmane.org>
2008-10-08 18:04             ` Trond Myklebust
2008-10-08 19:05               ` Talpey, Thomas
2008-10-08 15:49   ` [PATCH 13/15] RPC/RDMA: harden connection logic against missing/late rdma_cm upcalls Tom Talpey
2008-10-08 15:49   ` [PATCH 14/15] RPC/RDMA: reformat a debug printk to keep lines together Tom Talpey
2008-10-08 15:49   ` [PATCH 15/15] RPC/RDMA: optionally emit useful transport info upon connect/disconnect Tom Talpey

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=20081008154744.1336.20909.stgit@tmt3.nane.netapp.com \
    --to=talpey@netapp.com \
    --cc=linux-nfs@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