From patchwork Wed Feb 9 14:11:11 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "D. Wythe" X-Patchwork-Id: 12740315 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id F04A9C433F5 for ; Wed, 9 Feb 2022 14:11:36 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S234831AbiBIOLb (ORCPT ); Wed, 9 Feb 2022 09:11:31 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:45618 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S234789AbiBIOLU (ORCPT ); Wed, 9 Feb 2022 09:11:20 -0500 Received: from out30-43.freemail.mail.aliyun.com (out30-43.freemail.mail.aliyun.com [115.124.30.43]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 3722EC06157B; Wed, 9 Feb 2022 06:11:23 -0800 (PST) X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R171e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=e01e04400;MF=alibuda@linux.alibaba.com;NM=1;PH=DS;RN=7;SR=0;TI=SMTPD_---0V4.YhE9_1644415880; Received: from localhost(mailfrom:alibuda@linux.alibaba.com fp:SMTPD_---0V4.YhE9_1644415880) by smtp.aliyun-inc.com(127.0.0.1); Wed, 09 Feb 2022 22:11:20 +0800 From: "D. Wythe" To: kgraul@linux.ibm.com Cc: kuba@kernel.org, davem@davemloft.net, netdev@vger.kernel.org, linux-s390@vger.kernel.org, linux-rdma@vger.kernel.org, "D. Wythe" Subject: [PATCH net-next v6 1/5] net/smc: Make smc_tcp_listen_work() independent Date: Wed, 9 Feb 2022 22:11:11 +0800 Message-Id: <75b1108013d44bdaab905db9ab75ad2a16619b57.1644413637.git.alibuda@linux.alibaba.com> X-Mailer: git-send-email 1.8.3.1 In-Reply-To: References: Precedence: bulk List-ID: X-Mailing-List: linux-rdma@vger.kernel.org From: "D. Wythe" In multithread and 10K connections benchmark, the backend TCP connection established very slowly, and lots of TCP connections stay in SYN_SENT state. Client: smc_run wrk -c 10000 -t 4 http://server the netstate of server host shows like: 145042 times the listen queue of a socket overflowed 145042 SYNs to LISTEN sockets dropped One reason of this issue is that, since the smc_tcp_listen_work() shared the same workqueue (smc_hs_wq) with smc_listen_work(), while the smc_listen_work() do blocking wait for smc connection established. Once the workqueue became congested, it's will block the accept() from TCP listen. This patch creates a independent workqueue(smc_tcp_ls_wq) for smc_tcp_listen_work(), separate it from smc_listen_work(), which is quite acceptable considering that smc_tcp_listen_work() runs very fast. Signed-off-by: D. Wythe --- net/smc/af_smc.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 00b2e9d..4969ac8 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -59,6 +59,7 @@ * creation on client */ +static struct workqueue_struct *smc_tcp_ls_wq; /* wq for tcp listen work */ struct workqueue_struct *smc_hs_wq; /* wq for handshake work */ struct workqueue_struct *smc_close_wq; /* wq for close work */ @@ -2227,7 +2228,7 @@ static void smc_clcsock_data_ready(struct sock *listen_clcsock) lsmc->clcsk_data_ready(listen_clcsock); if (lsmc->sk.sk_state == SMC_LISTEN) { sock_hold(&lsmc->sk); /* sock_put in smc_tcp_listen_work() */ - if (!queue_work(smc_hs_wq, &lsmc->tcp_listen_work)) + if (!queue_work(smc_tcp_ls_wq, &lsmc->tcp_listen_work)) sock_put(&lsmc->sk); } } @@ -3024,9 +3025,14 @@ static int __init smc_init(void) goto out_nl; rc = -ENOMEM; + + smc_tcp_ls_wq = alloc_workqueue("smc_tcp_ls_wq", 0, 0); + if (!smc_tcp_ls_wq) + goto out_pnet; + smc_hs_wq = alloc_workqueue("smc_hs_wq", 0, 0); if (!smc_hs_wq) - goto out_pnet; + goto out_alloc_tcp_ls_wq; smc_close_wq = alloc_workqueue("smc_close_wq", 0, 0); if (!smc_close_wq) @@ -3097,6 +3103,8 @@ static int __init smc_init(void) destroy_workqueue(smc_close_wq); out_alloc_hs_wq: destroy_workqueue(smc_hs_wq); +out_alloc_tcp_ls_wq: + destroy_workqueue(smc_tcp_ls_wq); out_pnet: smc_pnet_exit(); out_nl: @@ -3115,6 +3123,7 @@ static void __exit smc_exit(void) smc_core_exit(); smc_ib_unregister_client(); destroy_workqueue(smc_close_wq); + destroy_workqueue(smc_tcp_ls_wq); destroy_workqueue(smc_hs_wq); proto_unregister(&smc_proto6); proto_unregister(&smc_proto); From patchwork Wed Feb 9 14:11:12 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "D. Wythe" X-Patchwork-Id: 12740317 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 0AB47C433F5 for ; Wed, 9 Feb 2022 14:11:39 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S234841AbiBIOLd (ORCPT ); Wed, 9 Feb 2022 09:11:33 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:46510 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S234823AbiBIOLb (ORCPT ); Wed, 9 Feb 2022 09:11:31 -0500 Received: from out30-45.freemail.mail.aliyun.com (out30-45.freemail.mail.aliyun.com [115.124.30.45]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 1D250C05CB97; Wed, 9 Feb 2022 06:11:26 -0800 (PST) X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R111e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=e01e04395;MF=alibuda@linux.alibaba.com;NM=1;PH=DS;RN=7;SR=0;TI=SMTPD_---0V4.ipeQ_1644415881; Received: from localhost(mailfrom:alibuda@linux.alibaba.com fp:SMTPD_---0V4.ipeQ_1644415881) by smtp.aliyun-inc.com(127.0.0.1); Wed, 09 Feb 2022 22:11:21 +0800 From: "D. Wythe" To: kgraul@linux.ibm.com Cc: kuba@kernel.org, davem@davemloft.net, netdev@vger.kernel.org, linux-s390@vger.kernel.org, linux-rdma@vger.kernel.org, "D. Wythe" Subject: [PATCH net-next v6 2/5] net/smc: Limit backlog connections Date: Wed, 9 Feb 2022 22:11:12 +0800 Message-Id: <412a5ddd496c5966a8910435a33552c78868d86d.1644413637.git.alibuda@linux.alibaba.com> X-Mailer: git-send-email 1.8.3.1 In-Reply-To: References: Precedence: bulk List-ID: X-Mailing-List: linux-rdma@vger.kernel.org From: "D. Wythe" Current implementation does not handling backlog semantics, one potential risk is that server will be flooded by infinite amount connections, even if client was SMC-incapable. This patch works to put a limit on backlog connections, referring to the TCP implementation, we divides SMC connections into two categories: 1. Half SMC connection, which includes all TCP established while SMC not connections. 2. Full SMC connection, which includes all SMC established connections. For half SMC connection, since all half SMC connections starts with TCP established, we can achieve our goal by put a limit before TCP established. Refer to the implementation of TCP, this limits will based on not only the half SMC connections but also the full connections, which is also a constraint on full SMC connections. For full SMC connections, although we know exactly where it starts, it's quite hard to put a limit before it. The easiest way is to block wait before receive SMC confirm CLC message, while it's under protection by smc_server_lgr_pending, a global lock, which leads this limit to the entire host instead of a single listen socket. Another way is to drop the full connections, but considering the cast of SMC connections, we prefer to keep full SMC connections. Even so, the limits of full SMC connections still exists, see commits about half SMC connection below. After this patch, the limits of backend connection shows like: For SMC: 1. Client with SMC-capability can makes 2 * backlog full SMC connections or 1 * backlog half SMC connections and 1 * backlog full SMC connections at most. 2. Client without SMC-capability can only makes 1 * backlog half TCP connections and 1 * backlog full TCP connections. Signed-off-by: D. Wythe --- net/smc/af_smc.c | 43 +++++++++++++++++++++++++++++++++++++++++++ net/smc/smc.h | 4 ++++ 2 files changed, 47 insertions(+) diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 4969ac8..ebfce3d 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -73,6 +73,34 @@ static void smc_set_keepalive(struct sock *sk, int val) smc->clcsock->sk->sk_prot->keepalive(smc->clcsock->sk, val); } +static struct sock *smc_tcp_syn_recv_sock(const struct sock *sk, struct sk_buff *skb, + struct request_sock *req, + struct dst_entry *dst, + struct request_sock *req_unhash, + bool *own_req) +{ + struct smc_sock *smc; + + smc = (struct smc_sock *)((uintptr_t)sk->sk_user_data & ~SK_USER_DATA_NOCOPY); + + if (READ_ONCE(sk->sk_ack_backlog) + atomic_read(&smc->smc_pendings) > + sk->sk_max_ack_backlog) + goto drop; + + if (sk_acceptq_is_full(&smc->sk)) { + NET_INC_STATS(sock_net(sk), LINUX_MIB_LISTENOVERFLOWS); + goto drop; + } + + /* passthrough to origin syn recv sock fct */ + return smc->ori_af_ops->syn_recv_sock(sk, skb, req, dst, req_unhash, own_req); + +drop: + dst_release(dst); + tcp_listendrop(sk); + return NULL; +} + static struct smc_hashinfo smc_v4_hashinfo = { .lock = __RW_LOCK_UNLOCKED(smc_v4_hashinfo.lock), }; @@ -1595,6 +1623,9 @@ static void smc_listen_out(struct smc_sock *new_smc) struct smc_sock *lsmc = new_smc->listen_smc; struct sock *newsmcsk = &new_smc->sk; + if (tcp_sk(new_smc->clcsock->sk)->syn_smc) + atomic_dec(&lsmc->smc_pendings); + if (lsmc->sk.sk_state == SMC_LISTEN) { lock_sock_nested(&lsmc->sk, SINGLE_DEPTH_NESTING); smc_accept_enqueue(&lsmc->sk, newsmcsk); @@ -2200,6 +2231,9 @@ static void smc_tcp_listen_work(struct work_struct *work) if (!new_smc) continue; + if (tcp_sk(new_smc->clcsock->sk)->syn_smc) + atomic_inc(&lsmc->smc_pendings); + new_smc->listen_smc = lsmc; new_smc->use_fallback = lsmc->use_fallback; new_smc->fallback_rsn = lsmc->fallback_rsn; @@ -2266,6 +2300,15 @@ static int smc_listen(struct socket *sock, int backlog) smc->clcsock->sk->sk_data_ready = smc_clcsock_data_ready; smc->clcsock->sk->sk_user_data = (void *)((uintptr_t)smc | SK_USER_DATA_NOCOPY); + + /* save origin ops */ + smc->ori_af_ops = inet_csk(smc->clcsock->sk)->icsk_af_ops; + + smc->af_ops = *smc->ori_af_ops; + smc->af_ops.syn_recv_sock = smc_tcp_syn_recv_sock; + + inet_csk(smc->clcsock->sk)->icsk_af_ops = &smc->af_ops; + rc = kernel_listen(smc->clcsock, backlog); if (rc) { smc->clcsock->sk->sk_data_ready = smc->clcsk_data_ready; diff --git a/net/smc/smc.h b/net/smc/smc.h index 37b2001..5e5e38d 100644 --- a/net/smc/smc.h +++ b/net/smc/smc.h @@ -252,6 +252,10 @@ struct smc_sock { /* smc sock container */ bool use_fallback; /* fallback to tcp */ int fallback_rsn; /* reason for fallback */ u32 peer_diagnosis; /* decline reason from peer */ + atomic_t smc_pendings; /* pending smc connections */ + struct inet_connection_sock_af_ops af_ops; + const struct inet_connection_sock_af_ops *ori_af_ops; + /* origin af ops */ int sockopt_defer_accept; /* sockopt TCP_DEFER_ACCEPT * value From patchwork Wed Feb 9 14:11:13 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "D. Wythe" X-Patchwork-Id: 12740316 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id C728EC43217 for ; Wed, 9 Feb 2022 14:11:37 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S234807AbiBIOLc (ORCPT ); Wed, 9 Feb 2022 09:11:32 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:45728 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S234817AbiBIOLb (ORCPT ); Wed, 9 Feb 2022 09:11:31 -0500 Received: from out30-54.freemail.mail.aliyun.com (out30-54.freemail.mail.aliyun.com [115.124.30.54]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 1CE2CC05CB90; Wed, 9 Feb 2022 06:11:26 -0800 (PST) X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R111e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=e01e04395;MF=alibuda@linux.alibaba.com;NM=1;PH=DS;RN=7;SR=0;TI=SMTPD_---0V4.oVI6_1644415882; Received: from localhost(mailfrom:alibuda@linux.alibaba.com fp:SMTPD_---0V4.oVI6_1644415882) by smtp.aliyun-inc.com(127.0.0.1); Wed, 09 Feb 2022 22:11:23 +0800 From: "D. Wythe" To: kgraul@linux.ibm.com Cc: kuba@kernel.org, davem@davemloft.net, netdev@vger.kernel.org, linux-s390@vger.kernel.org, linux-rdma@vger.kernel.org, "D. Wythe" Subject: [PATCH net-next v6 3/5] net/smc: Fallback when handshake workqueue congested Date: Wed, 9 Feb 2022 22:11:13 +0800 Message-Id: <87d49c573a15e13a26314316692fccca91741042.1644413637.git.alibuda@linux.alibaba.com> X-Mailer: git-send-email 1.8.3.1 In-Reply-To: References: Precedence: bulk List-ID: X-Mailing-List: linux-rdma@vger.kernel.org From: "D. Wythe" This patch intends to provide a mechanism to allow automatic fallback to TCP according to the pressure of SMC handshake process. At present, frequent visits will cause the incoming connections to be backlogged in SMC handshake queue, raise the connections established time. Which is quite unacceptable for those applications who base on short lived connections. There are two ways to implement this mechanism: 1. Fallback when TCP established. 2. Fallback before TCP established. In the first way, we need to wait and receive CLC messages that the client will potentially send, and then actively reply with a decline message, in a sense, which is also a sort of SMC handshake, affect the connections established time on its way. In the second way, the only problem is that we need to inject SMC logic into TCP when it is about to reply the incoming SYN, since we already do that, it's seems not a problem anymore. And advantage is obvious, few additional processes are required to complete the fallback. This patch use the second way. Link: https://lore.kernel.org/all/1641301961-59331-1-git-send-email-alibuda@linux.alibaba.com/ Signed-off-by: D. Wythe --- include/linux/tcp.h | 1 + net/ipv4/tcp_input.c | 3 ++- net/smc/af_smc.c | 18 ++++++++++++++++++ 3 files changed, 21 insertions(+), 1 deletion(-) diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 78b91bb..1c4ae5d 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -394,6 +394,7 @@ struct tcp_sock { bool is_mptcp; #endif #if IS_ENABLED(CONFIG_SMC) + bool (*smc_in_limited)(const struct sock *sk); bool syn_smc; /* SYN includes SMC */ #endif diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index af94a6d..e817ec6 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -6703,7 +6703,8 @@ static void tcp_openreq_init(struct request_sock *req, ireq->ir_num = ntohs(tcp_hdr(skb)->dest); ireq->ir_mark = inet_request_mark(sk, skb); #if IS_ENABLED(CONFIG_SMC) - ireq->smc_ok = rx_opt->smc_ok; + ireq->smc_ok = rx_opt->smc_ok && !(tcp_sk(sk)->smc_in_limited && + tcp_sk(sk)->smc_in_limited(sk)); #endif } diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index ebfce3d..8175f60 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -101,6 +101,22 @@ static struct sock *smc_tcp_syn_recv_sock(const struct sock *sk, struct sk_buff return NULL; } +static bool smc_is_in_limited(const struct sock *sk) +{ + const struct smc_sock *smc; + + smc = (const struct smc_sock *) + ((uintptr_t)sk->sk_user_data & ~SK_USER_DATA_NOCOPY); + + if (!smc) + return true; + + if (workqueue_congested(WORK_CPU_UNBOUND, smc_hs_wq)) + return true; + + return false; +} + static struct smc_hashinfo smc_v4_hashinfo = { .lock = __RW_LOCK_UNLOCKED(smc_v4_hashinfo.lock), }; @@ -2309,6 +2325,8 @@ static int smc_listen(struct socket *sock, int backlog) inet_csk(smc->clcsock->sk)->icsk_af_ops = &smc->af_ops; + tcp_sk(smc->clcsock->sk)->smc_in_limited = smc_is_in_limited; + rc = kernel_listen(smc->clcsock, backlog); if (rc) { smc->clcsock->sk->sk_data_ready = smc->clcsk_data_ready; From patchwork Wed Feb 9 14:11:14 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "D. Wythe" X-Patchwork-Id: 12740319 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 06992C3525A for ; Wed, 9 Feb 2022 14:11:40 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S234843AbiBIOLe (ORCPT ); Wed, 9 Feb 2022 09:11:34 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:46496 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S234837AbiBIOLb (ORCPT ); Wed, 9 Feb 2022 09:11:31 -0500 Received: from out30-132.freemail.mail.aliyun.com (out30-132.freemail.mail.aliyun.com [115.124.30.132]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 15421C05CBA4; Wed, 9 Feb 2022 06:11:28 -0800 (PST) X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R901e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=e01e01424;MF=alibuda@linux.alibaba.com;NM=1;PH=DS;RN=7;SR=0;TI=SMTPD_---0V4.ipev_1644415884; Received: from localhost(mailfrom:alibuda@linux.alibaba.com fp:SMTPD_---0V4.ipev_1644415884) by smtp.aliyun-inc.com(127.0.0.1); Wed, 09 Feb 2022 22:11:24 +0800 From: "D. Wythe" To: kgraul@linux.ibm.com Cc: kuba@kernel.org, davem@davemloft.net, netdev@vger.kernel.org, linux-s390@vger.kernel.org, linux-rdma@vger.kernel.org, "D. Wythe" Subject: [PATCH net-next v6 4/5] net/smc: Dynamic control auto fallback by socket options Date: Wed, 9 Feb 2022 22:11:14 +0800 Message-Id: <47e067deb658407bc68899ee9e6827f8159f11b4.1644413637.git.alibuda@linux.alibaba.com> X-Mailer: git-send-email 1.8.3.1 In-Reply-To: References: Precedence: bulk List-ID: X-Mailing-List: linux-rdma@vger.kernel.org From: "D. Wythe" This patch aims to add dynamic control for SMC auto fallback, since we don't have socket option level for SMC yet, which requires we need to implement it at the same time. This patch does the following: - add new socket option level: SOL_SMC. - add new SMC socket option: SMC_AUTO_FALLBACK. - provide getter/setter for SMC socket options. Signed-off-by: D. Wythe --- include/linux/socket.h | 1 + include/uapi/linux/smc.h | 4 +++ net/smc/af_smc.c | 69 +++++++++++++++++++++++++++++++++++++++++++++++- net/smc/smc.h | 1 + 4 files changed, 74 insertions(+), 1 deletion(-) diff --git a/include/linux/socket.h b/include/linux/socket.h index 8ef26d8..6f85f5d 100644 --- a/include/linux/socket.h +++ b/include/linux/socket.h @@ -366,6 +366,7 @@ struct ucred { #define SOL_XDP 283 #define SOL_MPTCP 284 #define SOL_MCTP 285 +#define SOL_SMC 286 /* IPX options */ #define IPX_TYPE 1 diff --git a/include/uapi/linux/smc.h b/include/uapi/linux/smc.h index 6c2874f..9f2cbf8 100644 --- a/include/uapi/linux/smc.h +++ b/include/uapi/linux/smc.h @@ -284,4 +284,8 @@ enum { __SMC_NLA_SEID_TABLE_MAX, SMC_NLA_SEID_TABLE_MAX = __SMC_NLA_SEID_TABLE_MAX - 1 }; + +/* SMC socket options */ +#define SMC_AUTO_FALLBACK 1 /* allow auto fallback to TCP */ + #endif /* _UAPI_LINUX_SMC_H */ diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 8175f60..c313561 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -2325,7 +2325,8 @@ static int smc_listen(struct socket *sock, int backlog) inet_csk(smc->clcsock->sk)->icsk_af_ops = &smc->af_ops; - tcp_sk(smc->clcsock->sk)->smc_in_limited = smc_is_in_limited; + if (smc->auto_fallback) + tcp_sk(smc->clcsock->sk)->smc_in_limited = smc_is_in_limited; rc = kernel_listen(smc->clcsock, backlog); if (rc) { @@ -2620,6 +2621,67 @@ static int smc_shutdown(struct socket *sock, int how) return rc ? rc : rc1; } +static int __smc_getsockopt(struct socket *sock, int level, int optname, + char __user *optval, int __user *optlen) +{ + struct smc_sock *smc; + int val, len; + + smc = smc_sk(sock->sk); + + if (get_user(len, optlen)) + return -EFAULT; + + len = min_t(int, len, sizeof(int)); + + if (len < 0) + return -EINVAL; + + switch (optname) { + case SMC_AUTO_FALLBACK: + val = smc->auto_fallback; + break; + default: + return -EOPNOTSUPP; + } + + if (put_user(len, optlen)) + return -EFAULT; + if (copy_to_user(optval, &val, len)) + return -EFAULT; + + return 0; +} + +static int __smc_setsockopt(struct socket *sock, int level, int optname, + sockptr_t optval, unsigned int optlen) +{ + struct sock *sk = sock->sk; + struct smc_sock *smc; + int val, rc; + + smc = smc_sk(sk); + + lock_sock(sk); + switch (optname) { + case SMC_AUTO_FALLBACK: + if (optlen < sizeof(int)) + return -EINVAL; + if (copy_from_sockptr(&val, optval, sizeof(int))) + return -EFAULT; + + smc->auto_fallback = !!val; + rc = 0; + break; + default: + rc = -EOPNOTSUPP; + break; + } + release_sock(sk); + + return rc; +} + static int smc_setsockopt(struct socket *sock, int level, int optname, sockptr_t optval, unsigned int optlen) { @@ -2629,6 +2691,8 @@ static int smc_setsockopt(struct socket *sock, int level, int optname, if (level == SOL_TCP && optname == TCP_ULP) return -EOPNOTSUPP; + else if (level == SOL_SMC) + return __smc_setsockopt(sock, level, optname, optval, optlen); smc = smc_sk(sk); @@ -2711,6 +2775,9 @@ static int smc_getsockopt(struct socket *sock, int level, int optname, struct smc_sock *smc; int rc; + if (level == SOL_SMC) + return __smc_getsockopt(sock, level, optname, optval, optlen); + smc = smc_sk(sock->sk); mutex_lock(&smc->clcsock_release_lock); if (!smc->clcsock) { diff --git a/net/smc/smc.h b/net/smc/smc.h index 5e5e38d..a0bdf75 100644 --- a/net/smc/smc.h +++ b/net/smc/smc.h @@ -249,6 +249,7 @@ struct smc_sock { /* smc sock container */ struct work_struct smc_listen_work;/* prepare new accept socket */ struct list_head accept_q; /* sockets to be accepted */ spinlock_t accept_q_lock; /* protects accept_q */ + bool auto_fallback; /* auto fallabck to tcp */ bool use_fallback; /* fallback to tcp */ int fallback_rsn; /* reason for fallback */ u32 peer_diagnosis; /* decline reason from peer */ From patchwork Wed Feb 9 14:11:15 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "D. Wythe" X-Patchwork-Id: 12740318 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 9A1F4C4167E for ; Wed, 9 Feb 2022 14:11:39 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S234792AbiBIOLd (ORCPT ); Wed, 9 Feb 2022 09:11:33 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:46516 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S234828AbiBIOLb (ORCPT ); Wed, 9 Feb 2022 09:11:31 -0500 Received: from out30-54.freemail.mail.aliyun.com (out30-54.freemail.mail.aliyun.com [115.124.30.54]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id E7C61C05CBA3; Wed, 9 Feb 2022 06:11:28 -0800 (PST) X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R211e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=e01e04400;MF=alibuda@linux.alibaba.com;NM=1;PH=DS;RN=7;SR=0;TI=SMTPD_---0V4.kjv9_1644415885; Received: from localhost(mailfrom:alibuda@linux.alibaba.com fp:SMTPD_---0V4.kjv9_1644415885) by smtp.aliyun-inc.com(127.0.0.1); Wed, 09 Feb 2022 22:11:26 +0800 From: "D. Wythe" To: kgraul@linux.ibm.com Cc: kuba@kernel.org, davem@davemloft.net, netdev@vger.kernel.org, linux-s390@vger.kernel.org, linux-rdma@vger.kernel.org, "D. Wythe" Subject: [PATCH net-next v6 5/5] net/smc: Add global configure for auto fallback by netlink Date: Wed, 9 Feb 2022 22:11:15 +0800 Message-Id: <64348e3dcd0b74ed638e895fa217d03df9bec854.1644413637.git.alibuda@linux.alibaba.com> X-Mailer: git-send-email 1.8.3.1 In-Reply-To: References: Precedence: bulk List-ID: X-Mailing-List: linux-rdma@vger.kernel.org From: "D. Wythe" Although we can control SMC auto fallback through socket options, which means that applications who need it must modify their code. It's quite troublesome for many existing applications. This patch modifies the global default value of auto fallback through netlink, providing a way to auto fallback without modifying any code for applications. Suggested-by: Tony Lu Signed-off-by: D. Wythe --- include/net/netns/smc.h | 2 ++ include/uapi/linux/smc.h | 11 +++++++++++ net/smc/af_smc.c | 41 +++++++++++++++++++++++++++++++++++++++++ net/smc/smc.h | 6 ++++++ net/smc/smc_netlink.c | 15 +++++++++++++++ net/smc/smc_pnet.c | 3 +++ 6 files changed, 78 insertions(+) diff --git a/include/net/netns/smc.h b/include/net/netns/smc.h index ea8a9cf..61646ff 100644 --- a/include/net/netns/smc.h +++ b/include/net/netns/smc.h @@ -12,5 +12,7 @@ struct netns_smc { /* protect fback_rsn */ struct mutex mutex_fback_rsn; struct smc_stats_rsn *fback_rsn; + + bool auto_fallback; /* whether allow auto_fallback */ }; #endif diff --git a/include/uapi/linux/smc.h b/include/uapi/linux/smc.h index 9f2cbf8..28a590d 100644 --- a/include/uapi/linux/smc.h +++ b/include/uapi/linux/smc.h @@ -59,6 +59,9 @@ enum { SMC_NETLINK_DUMP_SEID, SMC_NETLINK_ENABLE_SEID, SMC_NETLINK_DISABLE_SEID, + SMC_NETLINK_DUMP_AUTO_FALLBACK, + SMC_NETLINK_ENABLE_AUTO_FALLBACK, + SMC_NETLINK_DISABLE_AUTO_FALLBACK, }; /* SMC_GENL_FAMILY top level attributes */ @@ -285,6 +288,14 @@ enum { SMC_NLA_SEID_TABLE_MAX = __SMC_NLA_SEID_TABLE_MAX - 1 }; +/* SMC_NETLINK_AUTO_FALLBACK attributes */ +enum { + SMC_NLA_AUTO_FALLBACK_UNSPEC, + SMC_NLA_AUTO_FALLBACK_ENABLED, /* u8 */ + __SMC_NLA_AUTO_FALLBACK_MAX, + SMC_NLA_AUTO_FALLBACK_MAX = __SMC_NLA_AUTO_FALLBACK_MAX - 1 +}; + /* SMC socket options */ #define SMC_AUTO_FALLBACK 1 /* allow auto fallback to TCP */ diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index c313561..5e56281 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -66,6 +66,44 @@ static void smc_tcp_listen_work(struct work_struct *); static void smc_connect_work(struct work_struct *); +int smc_nl_dump_auto_fallback(struct sk_buff *skb, struct netlink_callback *cb) +{ + struct smc_nl_dmp_ctx *cb_ctx = smc_nl_dmp_ctx(cb); + void *hdr; + + if (cb_ctx->pos[0]) + goto out; + + hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, + &smc_gen_nl_family, NLM_F_MULTI, + SMC_NETLINK_DUMP_AUTO_FALLBACK); + if (!hdr) + return -ENOMEM; + + if (nla_put_u8(skb, SMC_NLA_AUTO_FALLBACK_ENABLED, sock_net(skb->sk)->smc.auto_fallback)) + goto err; + + genlmsg_end(skb, hdr); + cb_ctx->pos[0] = 1; +out: + return skb->len; +err: + genlmsg_cancel(skb, hdr); + return -EMSGSIZE; +} + +int smc_nl_enable_auto_fallback(struct sk_buff *skb, struct genl_info *info) +{ + sock_net(skb->sk)->smc.auto_fallback = true; + return 0; +} + +int smc_nl_disable_auto_fallback(struct sk_buff *skb, struct genl_info *info) +{ + sock_net(skb->sk)->smc.auto_fallback = false; + return 0; +} + static void smc_set_keepalive(struct sock *sk, int val) { struct smc_sock *smc = smc_sk(sk); @@ -3006,6 +3044,9 @@ static int __smc_create(struct net *net, struct socket *sock, int protocol, smc->use_fallback = false; /* assume rdma capability first */ smc->fallback_rsn = 0; + /* default behavior from auto_fallback */ + smc->auto_fallback = net->smc.auto_fallback; + rc = 0; if (!clcsock) { rc = sock_create_kern(net, family, SOCK_STREAM, IPPROTO_TCP, diff --git a/net/smc/smc.h b/net/smc/smc.h index a0bdf75..4cfbd61 100644 --- a/net/smc/smc.h +++ b/net/smc/smc.h @@ -14,6 +14,7 @@ #include #include #include /* __aligned */ +#include #include #include "smc_ib.h" @@ -336,4 +337,9 @@ void smc_fill_gid_list(struct smc_link_group *lgr, struct smc_gidlist *gidlist, struct smc_ib_device *known_dev, u8 *known_gid); +/* smc auto_fallback interface for netlink */ +int smc_nl_dump_auto_fallback(struct sk_buff *skb, struct netlink_callback *cb); +int smc_nl_enable_auto_fallback(struct sk_buff *skb, struct genl_info *info); +int smc_nl_disable_auto_fallback(struct sk_buff *skb, struct genl_info *info); + #endif /* __SMC_H */ diff --git a/net/smc/smc_netlink.c b/net/smc/smc_netlink.c index f13ab06..9f3abf7 100644 --- a/net/smc/smc_netlink.c +++ b/net/smc/smc_netlink.c @@ -111,6 +111,21 @@ .flags = GENL_ADMIN_PERM, .doit = smc_nl_disable_seid, }, + { + .cmd = SMC_NETLINK_DUMP_AUTO_FALLBACK, + /* can be retrieved by unprivileged users */ + .dumpit = smc_nl_dump_auto_fallback, + }, + { + .cmd = SMC_NETLINK_ENABLE_AUTO_FALLBACK, + .flags = GENL_ADMIN_PERM, + .doit = smc_nl_enable_auto_fallback, + }, + { + .cmd = SMC_NETLINK_DISABLE_AUTO_FALLBACK, + .flags = GENL_ADMIN_PERM, + .doit = smc_nl_disable_auto_fallback, + }, }; static const struct nla_policy smc_gen_nl_policy[2] = { diff --git a/net/smc/smc_pnet.c b/net/smc/smc_pnet.c index 291f148..07d7ee8 100644 --- a/net/smc/smc_pnet.c +++ b/net/smc/smc_pnet.c @@ -868,6 +868,9 @@ int smc_pnet_net_init(struct net *net) smc_pnet_create_pnetids_list(net); + /* disable auto fallback by default */ + net->smc.auto_fallback = 0; + return 0; }