From nobody Sun Feb 8 04:11:39 2026 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 12F5BC7EE23 for ; Wed, 7 Jun 2023 18:21:05 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231388AbjFGSVC (ORCPT ); Wed, 7 Jun 2023 14:21:02 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:36860 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232415AbjFGSUt (ORCPT ); Wed, 7 Jun 2023 14:20:49 -0400 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 93D66198B for ; Wed, 7 Jun 2023 11:20:04 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1686162003; 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=XqvSAbXVPa8XwKzNr0PvZXw2FsrUtjvMm73oMSig34E=; b=GVv0lUv+mi+vUq/sBMaukckmfGHxoogweveadMhrWZ3LpExnFMvF37ghL/qj1l+ooI04fJ y2m+YGcEsFUJS0Q9B8pOnM2uo54Q1RTfQP49NePpN2ZcW09lAZSGcARH18xr/43HScy16E qT2yw5L6ST/5cVeot4UTAAprUdwZfLs= Received: from mimecast-mx02.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-617-EoBIkxF_MvKXdSL0ouQGYQ-1; Wed, 07 Jun 2023 14:20:00 -0400 X-MC-Unique: EoBIkxF_MvKXdSL0ouQGYQ-1 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.rdu2.redhat.com [10.11.54.6]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id A04B23806704; Wed, 7 Jun 2023 18:19:59 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.42.28.182]) by smtp.corp.redhat.com (Postfix) with ESMTP id 249F82166B25; Wed, 7 Jun 2023 18:19:41 +0000 (UTC) From: David Howells To: netdev@vger.kernel.org, Linus Torvalds Cc: David Howells , Chuck Lever , Boris Pismenny , John Fastabend , Jakub Kicinski , "David S. Miller" , Eric Dumazet , Paolo Abeni , Willem de Bruijn , David Ahern , Matthew Wilcox , Jens Axboe , linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: [PATCH net-next v6 05/14] tls/sw: Use splice_eof() to flush Date: Wed, 7 Jun 2023 19:19:11 +0100 Message-ID: <20230607181920.2294972-6-dhowells@redhat.com> In-Reply-To: <20230607181920.2294972-1-dhowells@redhat.com> References: <20230607181920.2294972-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.6 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Allow splice to end a TLS record after prematurely ending a splice/sendfile due to getting an EOF condition (->splice_read() returned 0) after splice had called TLS with a sendmsg() with MSG_MORE set when the user didn't set MSG_MORE. Suggested-by: Linus Torvalds Link: https://lore.kernel.org/r/CAHk-=3Dwh=3DV579PDYvkpnTobCLGczbgxpMgGmmhq= iTyE34Cpi5Gg@mail.gmail.com/ Signed-off-by: David Howells Reviewed-by: Jakub Kicinski cc: Chuck Lever cc: Boris Pismenny cc: John Fastabend cc: Eric Dumazet cc: "David S. Miller" cc: Paolo Abeni cc: Jens Axboe cc: Matthew Wilcox cc: netdev@vger.kernel.org --- net/tls/tls.h | 1 + net/tls/tls_main.c | 2 ++ net/tls/tls_sw.c | 74 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 77 insertions(+) diff --git a/net/tls/tls.h b/net/tls/tls.h index 0672acab2773..4922668fefaa 100644 --- a/net/tls/tls.h +++ b/net/tls/tls.h @@ -97,6 +97,7 @@ void tls_update_rx_zc_capable(struct tls_context *tls_ctx= ); void tls_sw_strparser_arm(struct sock *sk, struct tls_context *ctx); void tls_sw_strparser_done(struct tls_context *tls_ctx); int tls_sw_sendmsg(struct sock *sk, struct msghdr *msg, size_t size); +void tls_sw_splice_eof(struct socket *sock); int tls_sw_sendpage_locked(struct sock *sk, struct page *page, int offset, size_t size, int flags); int tls_sw_sendpage(struct sock *sk, struct page *page, diff --git a/net/tls/tls_main.c b/net/tls/tls_main.c index e02a0d882ed3..82ec5c654f32 100644 --- a/net/tls/tls_main.c +++ b/net/tls/tls_main.c @@ -957,6 +957,7 @@ static void build_proto_ops(struct proto_ops ops[TLS_NU= M_CONFIG][TLS_NUM_CONFIG] ops[TLS_BASE][TLS_BASE] =3D *base; =20 ops[TLS_SW ][TLS_BASE] =3D ops[TLS_BASE][TLS_BASE]; + ops[TLS_SW ][TLS_BASE].splice_eof =3D tls_sw_splice_eof; ops[TLS_SW ][TLS_BASE].sendpage_locked =3D tls_sw_sendpage_locked; =20 ops[TLS_BASE][TLS_SW ] =3D ops[TLS_BASE][TLS_BASE]; @@ -1027,6 +1028,7 @@ static void build_protos(struct proto prot[TLS_NUM_CO= NFIG][TLS_NUM_CONFIG], =20 prot[TLS_SW][TLS_BASE] =3D prot[TLS_BASE][TLS_BASE]; prot[TLS_SW][TLS_BASE].sendmsg =3D tls_sw_sendmsg; + prot[TLS_SW][TLS_BASE].splice_eof =3D tls_sw_splice_eof; prot[TLS_SW][TLS_BASE].sendpage =3D tls_sw_sendpage; =20 prot[TLS_BASE][TLS_SW] =3D prot[TLS_BASE][TLS_BASE]; diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index 38acc27a0dd0..a2fb0256ff1c 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -1157,6 +1157,80 @@ int tls_sw_sendmsg(struct sock *sk, struct msghdr *m= sg, size_t size) return copied > 0 ? copied : ret; } =20 +/* + * Handle unexpected EOF during splice without SPLICE_F_MORE set. + */ +void tls_sw_splice_eof(struct socket *sock) +{ + struct sock *sk =3D sock->sk; + struct tls_context *tls_ctx =3D tls_get_ctx(sk); + struct tls_sw_context_tx *ctx =3D tls_sw_ctx_tx(tls_ctx); + struct tls_rec *rec; + struct sk_msg *msg_pl; + ssize_t copied =3D 0; + bool retrying =3D false; + int ret =3D 0; + int pending; + + if (!ctx->open_rec) + return; + + mutex_lock(&tls_ctx->tx_lock); + lock_sock(sk); + +retry: + rec =3D ctx->open_rec; + if (!rec) + goto unlock; + + msg_pl =3D &rec->msg_plaintext; + + /* Check the BPF advisor and perform transmission. */ + ret =3D bpf_exec_tx_verdict(msg_pl, sk, false, TLS_RECORD_TYPE_DATA, + &copied, 0); + switch (ret) { + case 0: + case -EAGAIN: + if (retrying) + goto unlock; + retrying =3D true; + goto retry; + case -EINPROGRESS: + break; + default: + goto unlock; + } + + /* Wait for pending encryptions to get completed */ + spin_lock_bh(&ctx->encrypt_compl_lock); + ctx->async_notify =3D true; + + pending =3D atomic_read(&ctx->encrypt_pending); + spin_unlock_bh(&ctx->encrypt_compl_lock); + if (pending) + crypto_wait_req(-EINPROGRESS, &ctx->async_wait); + else + reinit_completion(&ctx->async_wait.completion); + + /* There can be no concurrent accesses, since we have no pending + * encrypt operations + */ + WRITE_ONCE(ctx->async_notify, false); + + if (ctx->async_wait.err) + goto unlock; + + /* Transmit if any encryptions have completed */ + if (test_and_clear_bit(BIT_TX_SCHEDULED, &ctx->tx_bitmask)) { + cancel_delayed_work(&ctx->tx_work.work); + tls_tx_records(sk, 0); + } + +unlock: + release_sock(sk); + mutex_unlock(&tls_ctx->tx_lock); +} + static int tls_sw_do_sendpage(struct sock *sk, struct page *page, int offset, size_t size, int flags) {