From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-6.8 required=3.0 tests=DKIM_INVALID,DKIM_SIGNED, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH,MAILING_LIST_MULTI,SIGNED_OFF_BY, SPF_PASS autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 23F32C282DA for ; Fri, 1 Feb 2019 19:57:41 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id D4A91218AF for ; Fri, 1 Feb 2019 19:57:40 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=fail reason="signature verification failed" (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="kvg8o39c" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1729223AbfBAT5k (ORCPT ); Fri, 1 Feb 2019 14:57:40 -0500 Received: from mail-it1-f194.google.com ([209.85.166.194]:51741 "EHLO mail-it1-f194.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727097AbfBAT5j (ORCPT ); Fri, 1 Feb 2019 14:57:39 -0500 Received: by mail-it1-f194.google.com with SMTP id w18so11109577ite.1 for ; Fri, 01 Feb 2019 11:57:39 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=sender:subject:from:to:cc:date:message-id:in-reply-to:references :user-agent:mime-version:content-transfer-encoding; bh=aJWyu4yqiF6oEJf2GFLO3VcdAYqwCPRZxwDL723FfcI=; b=kvg8o39csWhq8v8fRYYK9ZV7JOHu+tFiuxj30YoFjZ8CmTL/R6nMJlkvZnQ6d0Y9Ck R9L8b1WjoakmatRCJ6Un4j80rZlgFaA4yTVcWWT7sWiJ+9lS2Eq1aNJJ5KLPhyzoVSA5 HkeVA8GOYqwSg8NzJjctH0efHiawpeBMUgkUzIn6AfEdVmedhrtNUS3QOQxhy3gC8gef W3viLCZx+RKjtTkzDat0T9CW8YILrSnkluqGu9cl6PUkYkbpA2AqbdTLu7YTdWuvwuRd ledzpYcE/fRevIWKEyULfiSN6uTrHTg0aLUGEgB1eEAmiTvXQK5zn+W5Sg0tpStLUrIG Io+w== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:sender:subject:from:to:cc:date:message-id :in-reply-to:references:user-agent:mime-version :content-transfer-encoding; bh=aJWyu4yqiF6oEJf2GFLO3VcdAYqwCPRZxwDL723FfcI=; b=EhW2MN7VFQyKhgO5ZCc4VhFZv7fVb4cRjkn8KUvucv/z4YA+cT7a701Sc87YsdT6pz QLSxpA+FWOeGmdrT4PzhByOephaDll96p/w7AKGuU0rP9X3C7ZsnDtrfhHc24qTevfM3 AkKs6jWTn1TscAq3ltfySnobxYTyP3rUGztnxnfiE9WK/ieLMHGcgqNEDs+YfGcsV5Z/ CB8d1rdzrvwPQKpYANBzyxMC2qFe7z8HAd+qVgmR6T5qSRveGSypeFyKQGHD9GvY8NKT ECzhmpIQaPj6usxqvUzxu+M+IxtoJWj1a8kJqEH3hzdBQNIWmOTKWDm+vsfeExmmQKbu AVlg== X-Gm-Message-State: AHQUAubPSiQRkHa02uv0uLFBkIlTfrnCh+QgvC4iJnKvjype7Wq0qxGs IUZr2gBjFLoQphgjWvNbjks= X-Google-Smtp-Source: AHgI3IZgO0+dxm3bLsXiRrzUvDpuCOcmQQ6X6ojcJIcdWCJmD+q9Hx1+DfRrUqMjmwXbII4IjIzTTQ== X-Received: by 2002:a24:c2c1:: with SMTP id i184mr2439160itg.117.1549051058685; Fri, 01 Feb 2019 11:57:38 -0800 (PST) Received: from gateway.1015granger.net (c-68-61-232-219.hsd1.mi.comcast.net. [68.61.232.219]) by smtp.gmail.com with ESMTPSA id t25sm3280720ioj.17.2019.02.01.11.57.37 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Fri, 01 Feb 2019 11:57:38 -0800 (PST) Received: from manet.1015granger.net (manet.1015granger.net [192.168.1.51]) by gateway.1015granger.net (8.14.7/8.14.7) with ESMTP id x11JvbIT008699; Fri, 1 Feb 2019 19:57:37 GMT Subject: [PATCH RFC 02/10] SUNRPC: Remove rpc_xprt::tsh_size From: Chuck Lever To: linux-nfs@vger.kernel.org Cc: simo@redhat.com Date: Fri, 01 Feb 2019 14:57:37 -0500 Message-ID: <20190201195737.11389.19493.stgit@manet.1015granger.net> In-Reply-To: <20190201195538.11389.96106.stgit@manet.1015granger.net> References: <20190201195538.11389.96106.stgit@manet.1015granger.net> User-Agent: StGit/0.17.1-dirty MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Sender: linux-nfs-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-nfs@vger.kernel.org tsh_size was added to accommodate transports that send a pre-amble before each RPC message. However, this assumes the pre-amble is fixed in size, which isn't true for some transports. That makes tsh_size not very generic. Also I'd like to make the estimation of RPC send and receive buffer sizes more precise. tsh_size doesn't currently appear to be accounted for at all by call_allocate. Therefore let's just remove the tsh_size concept, and make the only transports that have a non-zero tsh_size employ a direct approach. Signed-off-by: Chuck Lever --- include/linux/sunrpc/xprt.h | 7 -- net/sunrpc/auth_gss/auth_gss.c | 3 - net/sunrpc/clnt.c | 1 net/sunrpc/svc.c | 19 +----- net/sunrpc/xprtrdma/svc_rdma_backchannel.c | 1 net/sunrpc/xprtrdma/transport.c | 1 net/sunrpc/xprtsock.c | 91 ++++++++++++++++++---------- 7 files changed, 65 insertions(+), 58 deletions(-) diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index ad7e910..3a39154 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -196,8 +196,6 @@ struct rpc_xprt { size_t max_payload; /* largest RPC payload size, in bytes */ - unsigned int tsh_size; /* size of transport specific - header */ struct rpc_wait_queue binding; /* requests waiting on rpcbind */ struct rpc_wait_queue sending; /* requests waiting to send */ @@ -362,11 +360,6 @@ struct rpc_xprt * xprt_alloc(struct net *net, size_t size, unsigned int max_req); void xprt_free(struct rpc_xprt *); -static inline __be32 *xprt_skip_transport_header(struct rpc_xprt *xprt, __be32 *p) -{ - return p + xprt->tsh_size; -} - static inline int xprt_enable_swap(struct rpc_xprt *xprt) { diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c index a42672e..4b52e2b 100644 --- a/net/sunrpc/auth_gss/auth_gss.c +++ b/net/sunrpc/auth_gss/auth_gss.c @@ -1563,8 +1563,7 @@ static void gss_pipe_free(struct gss_pipe *p) /* We compute the checksum for the verifier over the xdr-encoded bytes * starting with the xid and ending at the end of the credential: */ - iov.iov_base = xprt_skip_transport_header(req->rq_xprt, - req->rq_snd_buf.head[0].iov_base); + iov.iov_base = req->rq_snd_buf.head[0].iov_base; iov.iov_len = (u8 *)p - (u8 *)iov.iov_base; xdr_buf_from_iov(&iov, &verf_buf); diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index d7ec613..c4203f6 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -2331,7 +2331,6 @@ void rpc_force_rebind(struct rpc_clnt *clnt) /* FIXME: check buffer size? */ - p = xprt_skip_transport_header(req->rq_xprt, p); *p++ = req->rq_xid; /* XID */ *p++ = htonl(RPC_CALL); /* CALL */ *p++ = htonl(RPC_VERSION); /* RPC version */ diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index e87ddb9..dbd1969 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -1145,17 +1145,6 @@ static __printf(2,3) void svc_printk(struct svc_rqst *rqstp, const char *fmt, .. #endif /* - * Setup response header for TCP, it has a 4B record length field. - */ -static void svc_tcp_prep_reply_hdr(struct svc_rqst *rqstp) -{ - struct kvec *resv = &rqstp->rq_res.head[0]; - - /* tcp needs a space for the record length... */ - svc_putnl(resv, 0); -} - -/* * Common routine for processing the RPC request. */ static int @@ -1182,10 +1171,6 @@ static void svc_tcp_prep_reply_hdr(struct svc_rqst *rqstp) set_bit(RQ_USEDEFERRAL, &rqstp->rq_flags); clear_bit(RQ_DROPME, &rqstp->rq_flags); - /* Setup reply header */ - if (rqstp->rq_prot == IPPROTO_TCP) - svc_tcp_prep_reply_hdr(rqstp); - svc_putu32(resv, rqstp->rq_xid); vers = svc_getnl(argv); @@ -1443,6 +1428,10 @@ static void svc_tcp_prep_reply_hdr(struct svc_rqst *rqstp) goto out_drop; } + /* Reserve space for the record marker */ + if (rqstp->rq_prot == IPPROTO_TCP) + svc_putnl(resv, 0); + /* Returns 1 for send, 0 for drop */ if (likely(svc_process_common(rqstp, argv, resv))) return svc_send(rqstp); diff --git a/net/sunrpc/xprtrdma/svc_rdma_backchannel.c b/net/sunrpc/xprtrdma/svc_rdma_backchannel.c index b908f2c..907464c 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_backchannel.c +++ b/net/sunrpc/xprtrdma/svc_rdma_backchannel.c @@ -304,7 +304,6 @@ static int svc_rdma_bc_sendto(struct svcxprt_rdma *rdma, xprt->idle_timeout = RPCRDMA_IDLE_DISC_TO; xprt->prot = XPRT_TRANSPORT_BC_RDMA; - xprt->tsh_size = 0; xprt->ops = &xprt_rdma_bc_procs; memcpy(&xprt->addr, args->dstaddr, args->addrlen); diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c index fbc171e..e7274dc 100644 --- a/net/sunrpc/xprtrdma/transport.c +++ b/net/sunrpc/xprtrdma/transport.c @@ -332,7 +332,6 @@ xprt->idle_timeout = RPCRDMA_IDLE_DISC_TO; xprt->resvport = 0; /* privileged port not needed */ - xprt->tsh_size = 0; /* RPC-RDMA handles framing */ xprt->ops = &xprt_rdma_procs; /* diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 7754aa3..ae09d85 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -696,6 +696,40 @@ static void xs_stream_data_receive_workfn(struct work_struct *work) #define XS_SENDMSG_FLAGS (MSG_DONTWAIT | MSG_NOSIGNAL) +/* Common case: + * - stream transport + * - sending from byte 0 of the message + * - the message is wholly contained in @xdr's head iovec + */ +static int xs_send_rm_and_kvec(struct socket *sock, struct xdr_buf *xdr, + unsigned int remainder) +{ + struct msghdr msg = { + .msg_flags = XS_SENDMSG_FLAGS | (remainder ? MSG_MORE : 0) + }; + rpc_fraghdr marker = cpu_to_be32(RPC_LAST_STREAM_FRAGMENT | + (u32)xdr->len); + struct kvec iov[2] = { + { + .iov_base = &marker, + .iov_len = sizeof(marker) + }, + { + .iov_base = xdr->head[0].iov_base, + .iov_len = xdr->head[0].iov_len + }, + }; + int ret; + + ret = kernel_sendmsg(sock, &msg, iov, 2, + iov[0].iov_len + iov[1].iov_len); + if (ret < 0) + return ret; + if (ret < iov[0].iov_len) + return -EPIPE; + return ret - iov[0].iov_len; +} + static int xs_send_kvec(struct socket *sock, struct sockaddr *addr, int addrlen, struct kvec *vec, unsigned int base, int more) { struct msghdr msg = { @@ -779,7 +813,11 @@ static int xs_sendpages(struct socket *sock, struct sockaddr *addr, int addrlen, if (base < xdr->head[0].iov_len || addr != NULL) { unsigned int len = xdr->head[0].iov_len - base; remainder -= len; - err = xs_send_kvec(sock, addr, addrlen, &xdr->head[0], base, remainder != 0); + if (!base && !addr) + err = xs_send_rm_and_kvec(sock, xdr, remainder); + else + err = xs_send_kvec(sock, addr, addrlen, &xdr->head[0], + base, remainder != 0); if (remainder == 0 || err != len) goto out; *sent_p += err; @@ -869,16 +907,6 @@ static int xs_nospace(struct rpc_rqst *req) return transport->xmit.offset != 0 && req->rq_bytes_sent == 0; } -/* - * Construct a stream transport record marker in @buf. - */ -static inline void xs_encode_stream_record_marker(struct xdr_buf *buf) -{ - u32 reclen = buf->len - sizeof(rpc_fraghdr); - rpc_fraghdr *base = buf->head[0].iov_base; - *base = cpu_to_be32(RPC_LAST_STREAM_FRAGMENT | reclen); -} - /** * xs_local_send_request - write an RPC request to an AF_LOCAL socket * @req: pointer to RPC request @@ -905,8 +933,6 @@ static int xs_local_send_request(struct rpc_rqst *req) return -ENOTCONN; } - xs_encode_stream_record_marker(&req->rq_snd_buf); - xs_pktdump("packet data:", req->rq_svec->iov_base, req->rq_svec->iov_len); @@ -1057,8 +1083,6 @@ static int xs_tcp_send_request(struct rpc_rqst *req) return -ENOTCONN; } - xs_encode_stream_record_marker(&req->rq_snd_buf); - xs_pktdump("packet data:", req->rq_svec->iov_base, req->rq_svec->iov_len); @@ -2534,26 +2558,35 @@ static int bc_sendto(struct rpc_rqst *req) { int len; struct xdr_buf *xbufp = &req->rq_snd_buf; - struct rpc_xprt *xprt = req->rq_xprt; struct sock_xprt *transport = - container_of(xprt, struct sock_xprt, xprt); - struct socket *sock = transport->sock; + container_of(req->rq_xprt, struct sock_xprt, xprt); unsigned long headoff; unsigned long tailoff; + struct page *tailpage; + struct msghdr msg = { + .msg_flags = MSG_MORE + }; + rpc_fraghdr marker = cpu_to_be32(RPC_LAST_STREAM_FRAGMENT | + (u32)xbufp->len); + struct kvec iov = { + .iov_base = &marker, + .iov_len = sizeof(marker), + }; - xs_encode_stream_record_marker(xbufp); + len = kernel_sendmsg(transport->sock, &msg, &iov, 1, iov.iov_len); + if (len != iov.iov_len) + return -EAGAIN; + tailpage = NULL; + if (xbufp->tail[0].iov_len) + tailpage = virt_to_page(xbufp->tail[0].iov_base); tailoff = (unsigned long)xbufp->tail[0].iov_base & ~PAGE_MASK; headoff = (unsigned long)xbufp->head[0].iov_base & ~PAGE_MASK; - len = svc_send_common(sock, xbufp, + len = svc_send_common(transport->sock, xbufp, virt_to_page(xbufp->head[0].iov_base), headoff, - xbufp->tail[0].iov_base, tailoff); - - if (len != xbufp->len) { - printk(KERN_NOTICE "Error sending entire callback!\n"); - len = -EAGAIN; - } - + tailpage, tailoff); + if (len != xbufp->len) + return -EAGAIN; return len; } @@ -2793,7 +2826,6 @@ static struct rpc_xprt *xs_setup_local(struct xprt_create *args) transport = container_of(xprt, struct sock_xprt, xprt); xprt->prot = 0; - xprt->tsh_size = sizeof(rpc_fraghdr) / sizeof(u32); xprt->max_payload = RPC_MAX_FRAGMENT_SIZE; xprt->bind_timeout = XS_BIND_TO; @@ -2862,7 +2894,6 @@ static struct rpc_xprt *xs_setup_udp(struct xprt_create *args) transport = container_of(xprt, struct sock_xprt, xprt); xprt->prot = IPPROTO_UDP; - xprt->tsh_size = 0; /* XXX: header size can vary due to auth type, IPv6, etc. */ xprt->max_payload = (1U << 16) - (MAX_HEADER << 3); @@ -2942,7 +2973,6 @@ static struct rpc_xprt *xs_setup_tcp(struct xprt_create *args) transport = container_of(xprt, struct sock_xprt, xprt); xprt->prot = IPPROTO_TCP; - xprt->tsh_size = sizeof(rpc_fraghdr) / sizeof(u32); xprt->max_payload = RPC_MAX_FRAGMENT_SIZE; xprt->bind_timeout = XS_BIND_TO; @@ -3015,7 +3045,6 @@ static struct rpc_xprt *xs_setup_bc_tcp(struct xprt_create *args) transport = container_of(xprt, struct sock_xprt, xprt); xprt->prot = IPPROTO_TCP; - xprt->tsh_size = sizeof(rpc_fraghdr) / sizeof(u32); xprt->max_payload = RPC_MAX_FRAGMENT_SIZE; xprt->timeout = &xs_tcp_default_timeout;