From nobody Sun Dec 14 22:12:05 2025 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 EA117C001DC for ; Fri, 23 Jun 2023 22:57:06 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232515AbjFWW5F (ORCPT ); Fri, 23 Jun 2023 18:57:05 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:37742 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231912AbjFWW5B (ORCPT ); Fri, 23 Jun 2023 18:57:01 -0400 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 8A103272D for ; Fri, 23 Jun 2023 15:55:44 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1687560943; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=W+/sbivnn4SSL91vpgGdKmYtGhKmLadJeuepui2ZFTU=; b=f2r1Qvw3q3oZm88gMz200soN+8EyRgKdgTbBHdw2PEf7wLV1rsluzrQIKj6PTAHpyKNQr8 1DUAPmZ760u0MF6W4QVxqs3aOxMpirR6CMmmJfq4AuM27rhMTmlsqbn31zoFhiMi8mizSn 1Spq5+yRyA9IIU34Zuvies3axLVXFAo= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-84-lshSpaI8Mri8O6LpAzQ_CA-1; Fri, 23 Jun 2023 18:55:38 -0400 X-MC-Unique: lshSpaI8Mri8O6LpAzQ_CA-1 Received: from smtp.corp.redhat.com (int-mx01.intmail.prod.int.rdu2.redhat.com [10.11.54.1]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 3B008809F8F; Fri, 23 Jun 2023 22:55:37 +0000 (UTC) Received: from warthog.procyon.org.com (unknown [10.42.28.4]) by smtp.corp.redhat.com (Postfix) with ESMTP id CF64940C2063; Fri, 23 Jun 2023 22:55:34 +0000 (UTC) From: David Howells To: netdev@vger.kernel.org Cc: David Howells , Alexander Duyck , "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , Willem de Bruijn , David Ahern , Matthew Wilcox , Jens Axboe , linux-mm@kvack.org, linux-kernel@vger.kernel.org, Sagi Grimberg , Willem de Bruijn , Keith Busch , Jens Axboe , Christoph Hellwig , Chaitanya Kulkarni , linux-nvme@lists.infradead.org Subject: [PATCH net-next v5 07/16] nvme-tcp: Use sendmsg(MSG_SPLICE_PAGES) rather then sendpage Date: Fri, 23 Jun 2023 23:55:04 +0100 Message-ID: <20230623225513.2732256-8-dhowells@redhat.com> In-Reply-To: <20230623225513.2732256-1-dhowells@redhat.com> References: <20230623225513.2732256-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.1 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" When transmitting data, call down into TCP using a sendmsg with MSG_SPLICE_PAGES instead of sendpage. Signed-off-by: David Howells Tested-by: Sagi Grimberg Acked-by: Willem de Bruijn cc: Keith Busch cc: Jens Axboe cc: Christoph Hellwig cc: Chaitanya Kulkarni cc: "David S. Miller" cc: Eric Dumazet cc: Jakub Kicinski cc: Paolo Abeni cc: Jens Axboe cc: Matthew Wilcox cc: linux-nvme@lists.infradead.org cc: netdev@vger.kernel.org --- Notes: ver #4) - Cancel MSG_SPLICE_PAGES if the page being sent fails sendpage_ok(). =20 ver #3) - Split nvme/host from nvme/target changes. =20 ver #2) - Wrap lines at 80. drivers/nvme/host/tcp.c | 49 +++++++++++++++++++++++------------------ 1 file changed, 27 insertions(+), 22 deletions(-) diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c index bf0230442d57..47ae17f16c05 100644 --- a/drivers/nvme/host/tcp.c +++ b/drivers/nvme/host/tcp.c @@ -997,25 +997,28 @@ static int nvme_tcp_try_send_data(struct nvme_tcp_req= uest *req) u32 h2cdata_left =3D req->h2cdata_left; =20 while (true) { + struct bio_vec bvec; + struct msghdr msg =3D { + .msg_flags =3D MSG_DONTWAIT | MSG_SPLICE_PAGES, + }; struct page *page =3D nvme_tcp_req_cur_page(req); size_t offset =3D nvme_tcp_req_cur_offset(req); size_t len =3D nvme_tcp_req_cur_length(req); bool last =3D nvme_tcp_pdu_last_send(req, len); int req_data_sent =3D req->data_sent; - int ret, flags =3D MSG_DONTWAIT; + int ret; =20 if (last && !queue->data_digest && !nvme_tcp_queue_more(queue)) - flags |=3D MSG_EOR; + msg.msg_flags |=3D MSG_EOR; else - flags |=3D MSG_MORE | MSG_SENDPAGE_NOTLAST; + msg.msg_flags |=3D MSG_MORE; =20 - if (sendpage_ok(page)) { - ret =3D kernel_sendpage(queue->sock, page, offset, len, - flags); - } else { - ret =3D sock_no_sendpage(queue->sock, page, offset, len, - flags); - } + if (!sendpage_ok(page)) + msg.msg_flags &=3D ~MSG_SPLICE_PAGES, + + bvec_set_page(&bvec, page, len, offset); + iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, &bvec, 1, len); + ret =3D sock_sendmsg(queue->sock, &msg); if (ret <=3D 0) return ret; =20 @@ -1054,22 +1057,24 @@ static int nvme_tcp_try_send_cmd_pdu(struct nvme_tc= p_request *req) { struct nvme_tcp_queue *queue =3D req->queue; struct nvme_tcp_cmd_pdu *pdu =3D nvme_tcp_req_cmd_pdu(req); + struct bio_vec bvec; + struct msghdr msg =3D { .msg_flags =3D MSG_DONTWAIT | MSG_SPLICE_PAGES, }; bool inline_data =3D nvme_tcp_has_inline_data(req); u8 hdgst =3D nvme_tcp_hdgst_len(queue); int len =3D sizeof(*pdu) + hdgst - req->offset; - int flags =3D MSG_DONTWAIT; int ret; =20 if (inline_data || nvme_tcp_queue_more(queue)) - flags |=3D MSG_MORE | MSG_SENDPAGE_NOTLAST; + msg.msg_flags |=3D MSG_MORE; else - flags |=3D MSG_EOR; + msg.msg_flags |=3D MSG_EOR; =20 if (queue->hdr_digest && !req->offset) nvme_tcp_hdgst(queue->snd_hash, pdu, sizeof(*pdu)); =20 - ret =3D kernel_sendpage(queue->sock, virt_to_page(pdu), - offset_in_page(pdu) + req->offset, len, flags); + bvec_set_virt(&bvec, (void *)pdu + req->offset, len); + iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, &bvec, 1, len); + ret =3D sock_sendmsg(queue->sock, &msg); if (unlikely(ret <=3D 0)) return ret; =20 @@ -1093,6 +1098,8 @@ static int nvme_tcp_try_send_data_pdu(struct nvme_tcp= _request *req) { struct nvme_tcp_queue *queue =3D req->queue; struct nvme_tcp_data_pdu *pdu =3D nvme_tcp_req_data_pdu(req); + struct bio_vec bvec; + struct msghdr msg =3D { .msg_flags =3D MSG_DONTWAIT | MSG_MORE, }; u8 hdgst =3D nvme_tcp_hdgst_len(queue); int len =3D sizeof(*pdu) - req->offset + hdgst; int ret; @@ -1101,13 +1108,11 @@ static int nvme_tcp_try_send_data_pdu(struct nvme_t= cp_request *req) nvme_tcp_hdgst(queue->snd_hash, pdu, sizeof(*pdu)); =20 if (!req->h2cdata_left) - ret =3D kernel_sendpage(queue->sock, virt_to_page(pdu), - offset_in_page(pdu) + req->offset, len, - MSG_DONTWAIT | MSG_MORE | MSG_SENDPAGE_NOTLAST); - else - ret =3D sock_no_sendpage(queue->sock, virt_to_page(pdu), - offset_in_page(pdu) + req->offset, len, - MSG_DONTWAIT | MSG_MORE); + msg.msg_flags |=3D MSG_SPLICE_PAGES; + + bvec_set_virt(&bvec, (void *)pdu + req->offset, len); + iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, &bvec, 1, len); + ret =3D sock_sendmsg(queue->sock, &msg); if (unlikely(ret <=3D 0)) return ret;