Return-Path: <linux-nfs-owner@vger.kernel.org>
Received: from userp1040.oracle.com ([156.151.31.81]:36602 "EHLO
        userp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1751371AbdA3UcW (ORCPT
        <rfc822;linux-nfs@vger.kernel.org>); Mon, 30 Jan 2017 15:32:22 -0500
Content-Type: text/plain; charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 9.3 \(3124\))
Subject: Re: [PATCH v1 7/7] sunrpc: Allow keepalive ping on a credit-full transport
From: Chuck Lever <chuck.lever@oracle.com>
In-Reply-To: <5268b237-2c33-2058-d4b2-798afe021243@gmail.com>
Date: Mon, 30 Jan 2017 15:32:18 -0500
Cc: linux-rdma@vger.kernel.org,
        Linux NFS Mailing List <linux-nfs@vger.kernel.org>
Message-Id: <7694B03F-328F-472A-8A41-F3D6DCF0125E@oracle.com>
References: <20170126174806.5794.14678.stgit@manet.1015granger.net> <20170126175628.5794.56813.stgit@manet.1015granger.net> <5268b237-2c33-2058-d4b2-798afe021243@gmail.com>
To: Anna Schumaker <schumaker.anna@gmail.com>
Sender: linux-nfs-owner@vger.kernel.org
List-ID: <linux-nfs.vger.kernel.org>


> On Jan 30, 2017, at 3:29 PM, Anna Schumaker <schumaker.anna@gmail.com> wrote:
> 
> Hi Chuck,
> 
> On 01/26/2017 12:56 PM, Chuck Lever wrote:
>> Allow RPC-over-RDMA to send NULL pings even when the transport has
>> hit its credit limit.
>> 
>> One credit is reserved. It may be used only to send a keepalive
>> ping.
>> 
>> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
>> ---
>> include/linux/sunrpc/sched.h    |    2 ++
>> net/sunrpc/xprt.c               |    4 ++++
>> net/sunrpc/xprtrdma/transport.c |    4 +++-
>> net/sunrpc/xprtrdma/verbs.c     |   13 ++++++++-----
>> 4 files changed, 17 insertions(+), 6 deletions(-)
>> 
>> diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h
>> index 7ba040c..a65278e 100644
>> --- a/include/linux/sunrpc/sched.h
>> +++ b/include/linux/sunrpc/sched.h
>> @@ -127,6 +127,7 @@ struct rpc_task_setup {
>> #define RPC_TASK_TIMEOUT	0x1000		/* fail with ETIMEDOUT on timeout */
>> #define RPC_TASK_NOCONNECT	0x2000		/* return ENOTCONN if not connected */
>> #define RPC_TASK_NO_RETRANS_TIMEOUT	0x4000		/* wait forever for a reply */
>> +#define RPC_TASK_PRIORITY	0x8000		/* skip congestion control */
> 
> This file also defines various RPC_PRIORITY_<whatever> values that get used for the task->tk_priority field.  I wonder if there is a better name for this flag that would help keep people from confusing the two later down the line, but I'm also struggling to come up with anything better.

PRIORITY was a (lead) trial balloon, and also a sign of weak imagination.

The point of the flag is to bypass congestion control. RPC_TASK_SKIPCONG ?
Meh.


> Thoughts?
> Anna
> 
>> 
>> #define RPC_IS_ASYNC(t)		((t)->tk_flags & RPC_TASK_ASYNC)
>> #define RPC_IS_SWAPPER(t)	((t)->tk_flags & RPC_TASK_SWAPPER)
>> @@ -135,6 +136,7 @@ struct rpc_task_setup {
>> #define RPC_IS_SOFT(t)		((t)->tk_flags & (RPC_TASK_SOFT|RPC_TASK_TIMEOUT))
>> #define RPC_IS_SOFTCONN(t)	((t)->tk_flags & RPC_TASK_SOFTCONN)
>> #define RPC_WAS_SENT(t)		((t)->tk_flags & RPC_TASK_SENT)
>> +#define RPC_HAS_PRIORITY(t)	((t)->tk_flags & RPC_TASK_PRIORITY)
>> 
>> #define RPC_TASK_RUNNING	0
>> #define RPC_TASK_QUEUED		1
>> diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
>> index b530a28..e8b7b9f 100644
>> --- a/net/sunrpc/xprt.c
>> +++ b/net/sunrpc/xprt.c
>> @@ -392,6 +392,10 @@ static inline void xprt_release_write(struct rpc_xprt *xprt, struct rpc_task *ta
>> {
>> 	struct rpc_rqst *req = task->tk_rqstp;
>> 
>> +	if (RPC_HAS_PRIORITY(task)) {
>> +		req->rq_cong = 0;
>> +		return 1;
>> +	}
>> 	if (req->rq_cong)
>> 		return 1;
>> 	dprintk("RPC: %5u xprt_cwnd_limited cong = %lu cwnd = %lu\n",
>> diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c
>> index f97c851..e43cc85 100644
>> --- a/net/sunrpc/xprtrdma/transport.c
>> +++ b/net/sunrpc/xprtrdma/transport.c
>> @@ -538,7 +538,9 @@ static void rpcrdma_keepalive_release(void *calldata)
>> 
>> 	data = xprt_get(xprt);
>> 	null_task = rpc_call_null_helper(task->tk_client, xprt, NULL,
>> -					 RPC_TASK_SOFT | RPC_TASK_ASYNC,
>> +					 RPC_TASK_SOFT |
>> +					 RPC_TASK_ASYNC |
>> +					 RPC_TASK_PRIORITY,
>> 					 &rpcrdma_keepalive_call_ops, data);
>> 	if (!IS_ERR(null_task))
>> 		rpc_put_task(null_task);
>> diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
>> index 7e14b73..3e218dd 100644
>> --- a/net/sunrpc/xprtrdma/verbs.c
>> +++ b/net/sunrpc/xprtrdma/verbs.c
>> @@ -136,19 +136,20 @@
>> static void
>> rpcrdma_update_granted_credits(struct rpcrdma_rep *rep)
>> {
>> -	struct rpcrdma_msg *rmsgp = rdmab_to_msg(rep->rr_rdmabuf);
>> 	struct rpcrdma_buffer *buffer = &rep->rr_rxprt->rx_buf;
>> +	__be32 *p = rep->rr_rdmabuf->rg_base;
>> 	u32 credits;
>> 
>> 	if (rep->rr_len < RPCRDMA_HDRLEN_ERR)
>> 		return;
>> 
>> -	credits = be32_to_cpu(rmsgp->rm_credit);
>> +	credits = be32_to_cpup(p + 2);
>> +	if (credits > buffer->rb_max_requests)
>> +		credits = buffer->rb_max_requests;
>> +	/* Reserve one credit for keepalive ping */
>> +	credits--;
>> 	if (credits == 0)
>> 		credits = 1;	/* don't deadlock */
>> -	else if (credits > buffer->rb_max_requests)
>> -		credits = buffer->rb_max_requests;
>> -
>> 	atomic_set(&buffer->rb_credits, credits);
>> }
>> 
>> @@ -914,6 +915,8 @@ struct rpcrdma_rep *
>> 	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
>> 	int i, rc;
>> 
>> +	if (r_xprt->rx_data.max_requests < 2)
>> +		return -EINVAL;
>> 	buf->rb_max_requests = r_xprt->rx_data.max_requests;
>> 	buf->rb_bc_srv_max_requests = 0;
>> 	atomic_set(&buf->rb_credits, 1);
>> 
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

--
Chuck Lever