From patchwork Tue Sep 15 15:23:54 2026 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ralf Lici X-Patchwork-Id: 5344 Return-Path: Delivered-To: patchwork@openvpn.net Received: by 2002:a05:7000:6446:b0:8a0:ea1f:253a with SMTP id n6csp5063811mag; Tue, 15 Sep 2026 08:24:23 -0700 (PDT) X-Forwarded-Encrypted: i=2; AKwUvBwYbq7whhY9N/GWthtucHpFyROLe98NZOt1sgoHJEz3lcKKvBo4bGCj45OMPx/oCKYf3bIPTejHuAA=@openvpn.net X-Received: by 2002:a05:6870:720c:b0:47c:1cfe:5cd0 with SMTP id 586e51a60fabf-481f5144002mr9694630fac.0.1789485863020; Tue, 15 Sep 2026 08:24:23 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1789485863; cv=none; d=google.com; s=arc-20260327; b=SgwxhqMZGDZiNPHQFIjAxkKri7sl0yxisKKipQB8hJKkJJzYZr1H4ylJLy41FWxYTC XR2ho31yz78197OT5+yrfSlE4RLtDAvnJv+Vs/xl0w42kebkvx5BccxD9AOy+lBgTF2x 81HO4Kiu8eq8TFwnJKxGJPdaFOI5uN6578zcYCWT4zbSVTsroYvWURE3vPp1QX+Ord1r rLANUCVqDhayDWuancStDGZ25dY9H5EGVEZ1SuTBKuCmVBDLT+LL4HfhvTCueufuYSYl dZb2uEiJp1bCZeWzL4c0Vh3UHOUrIPhX3rP6vAhFVrCB8m6PnLPcZ83oGSTTfsE3ZtDB wrzw== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20260327; h=errors-to:content-transfer-encoding:list-subscribe:list-help :list-post:list-archive:list-unsubscribe:list-id:precedence:subject :mime-version:references:in-reply-to:message-id:date:to:from :dkim-signature:dkim-signature:dkim-signature:dkim-signature; bh=jMHjxBkKR1Pn2eSVg8SkFx3NnTrEHwm+LfaSL3OEnV0=; fh=4NbAC/LsuMLI0S0hprUlLSLCiHwg6SCAifhH718Jh0Q=; b=ZXP/kSe0gauTbcl5g11jDmo9JAIbE3cZyYVA+xVmaXsr19g0k+eBsteNFoO9VpkN1U W+Y2NhFBJpAh1yRWxSllfNm2A2nrOnq/tkoMswYchJN2qwNrlaQg9yKTI6OZTmiS8VDA E/3pJ+NpLiHXTL7TPH39KvxrsveTYHA5pvyS3oFDfFLgE5w+O8n9vWBWqdkh8a6o1dVG e6amH5jmcRXd8HSJ4GEeoYu6SeGBzviaTCLGsUkH0BVaK7OK2nBRuTMHCauN88DFWJlT Nu9/BwCfnDgdSepDNxK95dlwad0RHBkaXL0oyhwkqMPJTfPPTMsAR21dbtcMjUXmi1SV C7Hg==; dara=google.com ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@lists.sourceforge.net header.s=beta header.b=HQtMVXdi; dkim=neutral (body hash did not verify) header.i=@sourceforge.net header.s=x header.b=mdnnDM3o; dkim=neutral (body hash did not verify) header.i=@sf.net header.s=x header.b=crqgyZdA; dkim=neutral (body hash did not verify) header.i=@mandelbit.com header.s=MBO0001 header.b=iKExSTCK; spf=pass (google.com: domain of openvpn-devel-bounces@lists.sourceforge.net designates 216.105.38.7 as permitted sender) smtp.mailfrom=openvpn-devel-bounces@lists.sourceforge.net Received: from lists.sourceforge.net (lists.sourceforge.net. [216.105.38.7]) by mx.google.com with ESMTPS id 586e51a60fabf-47df89c0fcesi11296985fac.82.2026.09.15.08.24.21 (version=TLS1_2 cipher=ECDHE-ECDSA-AES128-GCM-SHA256 bits=128/128); Tue, 15 Sep 2026 08:24:22 -0700 (PDT) Received-SPF: pass (google.com: domain of openvpn-devel-bounces@lists.sourceforge.net designates 216.105.38.7 as permitted sender) client-ip=216.105.38.7; Authentication-Results: mx.google.com; dkim=pass header.i=@lists.sourceforge.net header.s=beta header.b=HQtMVXdi; dkim=neutral (body hash did not verify) header.i=@sourceforge.net header.s=x header.b=mdnnDM3o; dkim=neutral (body hash did not verify) header.i=@sf.net header.s=x header.b=crqgyZdA; dkim=neutral (body hash did not verify) header.i=@mandelbit.com header.s=MBO0001 header.b=iKExSTCK; spf=pass (google.com: domain of openvpn-devel-bounces@lists.sourceforge.net designates 216.105.38.7 as permitted sender) smtp.mailfrom=openvpn-devel-bounces@lists.sourceforge.net DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.sourceforge.net; s=beta; h=Content-Transfer-Encoding:Content-Type: List-Subscribe:List-Help:List-Post:List-Archive:List-Unsubscribe:List-Id: Subject:MIME-Version:References:In-Reply-To:Message-ID:Date:To:From:Sender: Reply-To:Cc:Content-ID:Content-Description:Resent-Date:Resent-From: Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Owner; bh=jMHjxBkKR1Pn2eSVg8SkFx3NnTrEHwm+LfaSL3OEnV0=; b=HQtMVXdirUT/Z8x8YHbSeGpzAy Js7TF0YBGn9nhyjD/jBPWBPX3+mpmDgpQ+D+XdQiyWkKxEFyqXKdkIEMsbpr/DryowuGNv8gkcRvR CeFQD80169ko9/2bG2ygeWfzZ+qv8MqRyNX+n5lkdP6YCN0qiq459IevdK7XWInntjXw=; Received: from [127.0.0.1] (helo=sfs-ml-2.v29.lw.sourceforge.com) by sfs-ml-2.v29.lw.sourceforge.com with esmtp (Exim 4.95) (envelope-from ) id 1x6V16-0000JT-V8; Tue, 15 Sep 2026 15:24:17 +0000 Received: from [172.30.29.66] (helo=mx.sourceforge.net) by sfs-ml-2.v29.lw.sourceforge.com with esmtps (TLS1.2) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.95) (envelope-from ) id 1x6V13-0000Ik-Iz for openvpn-devel@lists.sourceforge.net; Tue, 15 Sep 2026 15:24:14 +0000 DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=sourceforge.net; s=x; h=Content-Transfer-Encoding:MIME-Version:References: In-Reply-To:Message-ID:Date:Subject:To:From:Sender:Reply-To:Cc:Content-Type: Content-ID:Content-Description:Resent-Date:Resent-From:Resent-Sender: Resent-To:Resent-Cc:Resent-Message-ID:List-Id:List-Help:List-Unsubscribe: List-Subscribe:List-Post:List-Owner:List-Archive; bh=Kut5kWrtVSzjUahQOQL0CLcLzWSFwxNf6jCq+jiq2QE=; b=mdnnDM3o1CKAmFmsG1R+xr+Kb1 i3Psjs3j5rYF9jJWsfSoqQ9qEAKUgaVweAO5wc3WuEKUOq4AqraN+LFVTgf7pyH2I4znrwwNNEvjT ipO8yHdqkXYsKbM+RFpW9eO7sAf7TvM9SaDKyW86NQ29DUS1D/CRrAwtnNm9khSXJDZY=; DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=sf.net; s=x ; h=Content-Transfer-Encoding:MIME-Version:References:In-Reply-To:Message-ID: Date:Subject:To:From:Sender:Reply-To:Cc:Content-Type:Content-ID: Content-Description:Resent-Date:Resent-From:Resent-Sender:Resent-To:Resent-Cc :Resent-Message-ID:List-Id:List-Help:List-Unsubscribe:List-Subscribe: List-Post:List-Owner:List-Archive; bh=Kut5kWrtVSzjUahQOQL0CLcLzWSFwxNf6jCq+jiq2QE=; b=crqgyZdA500QJs04MaI9psZfSc LDtLSQiw/hkENY/3yxFjI0UHE//lVQ/AEKKceYWA4dyfSvnpA1ss70+PgW71QskumBl9+VE+FER/2 ooWHPnn5OLB9D8oyH/FOQpYvmWqB2rXpzt2vlRk/jfwX0AhXKjaJVhvKxXRI7iOavd0U=; Received: from mout-b-112.mailbox.org ([195.10.208.42]) by sfi-mx-1.v28.lw.sourceforge.com with esmtps (TLS1.2:ECDHE-RSA-AES256-GCM-SHA384:256) (Exim 4.95) id 1x6V12-0003bS-9q for openvpn-devel@lists.sourceforge.net; Tue, 15 Sep 2026 15:24:14 +0000 Received: from smtp102.mailbox.org (smtp102.mailbox.org [IPv6:2001:67c:2050:b231:465::102]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange x25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (No client certificate requested) by mout-b-112.mailbox.org (Postfix) with ESMTPS id 4hkm5K04FGz5vT4 for ; Tue, 15 Sep 2026 17:24:05 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=mandelbit.com; s=MBO0001; t=1789485845; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=Kut5kWrtVSzjUahQOQL0CLcLzWSFwxNf6jCq+jiq2QE=; b=iKExSTCKHi0V3Zwtbr9u7YYhO38J8u3y0aa/x7zVqq+nWWn2lZGZ7Hu/PW9X6yNk2Gs513 jnWkBNxAd3g0I6STK/y9K4D8F+SdxgT51EuywFJyiVtkGjszmYnlTBMLmtMYQmXpBAvP9n BP4Lc+eKUAk8Qbh/P0zNERF4h0LKW1+dorZt3FbpMrE8NIz82DZ5e8uZKGyn70fUpIbohN HvsjgwduL0oFXB0iU41GVjwwNdYmH43MhyLGRDbkKkyIDdv/XeY2hUJNUkuVSX2XfYoRKP +kYRSB1ZIlf43Z3w+/uLIINkwnizv4iSg3nNkaAApATgatJMgu9fgMDQZOBXRQ== Authentication-Results: outgoing_mbo_mout; dkim=none; spf=pass (outgoing_mbo_mout: domain of ralf@mandelbit.com designates 2001:67c:2050:b231:465::102 as permitted sender) smtp.mailfrom=ralf@mandelbit.com From: Ralf Lici To: openvpn-devel@lists.sourceforge.net Date: Tue, 15 Sep 2026 17:23:54 +0200 Message-ID: <712708baf265c5509aed4f9d476abf514a242b8a.1789485693.git.ralf@mandelbit.com> In-Reply-To: References: MIME-Version: 1.0 X-Rspamd-Queue-Id: 4hkm5K04FGz5vT4 X-Spam-Score: -0.2 (/) X-Spam-Report: Spam detection software, running on the system "sfi-spamd-2.hosts.colo.sdot.me", has NOT identified this incoming email as spam. The original message has been attached to this so you can view it or label similar future email. If you have any questions, see the administrator of that system for details. Content preview: Frag-list GRO exposes later encrypted records before the receive path decrypts the current one. Use this lookahead to request write ownership of linear ciphertext cache lines two records in advance (a [...] Content analysis details: (-0.2 points, 5.0 required) pts rule name description ---- ---------------------- -------------------------------------------------- -0.1 DKIM_VALID_EF Message has a valid DKIM or DK signature from envelope-from domain -0.1 DKIM_VALID Message has at least one valid DKIM or DK signature -0.1 DKIM_VALID_AU Message has a valid DKIM or DK signature from author's domain 0.1 DKIM_SIGNED Message has a DKIM or DK signature, not necessarily valid X-Headers-End: 1x6V12-0003bS-9q Subject: [Openvpn-devel] [RFC ovpn net-next 6/9] ovpn: prefetch encrypted records before GRO batch decryption X-BeenThere: openvpn-devel@lists.sourceforge.net X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: openvpn-devel-bounces@lists.sourceforge.net X-getmail-retrieved-from-mailbox: Inbox X-GMAIL-THRID: 1876411928268645580 X-GMAIL-MSGID: 1876411928268645580 Frag-list GRO exposes later encrypted records before the receive path decrypts the current one. Use this lookahead to request write ownership of linear ciphertext cache lines two records in advance (and maintain the same distance throughout the batch), overlapping their memory access latency with the current AEAD operation. An ordinary single-record UDP receive has no later record and therefore skips the prefetch path. Only prefetch the linear part of each skb. Walking non-linear fragments here would duplicate the scatterlist walk performed by crypto and could cost more than the cache hint saves. A same-binary comparison using three 30-second samples per direction found distances one and two effectively tied forward, while distance two was 3.3% faster reverse and less variable in both directions. Profiling also measured slightly fewer decrypt cycles at distance two than at one, while wider distances provided no repeatable benefit. On a direct 100 Gbit/s ConnectX-5 link using one TCP stream, AES-128-GCM, a 1408-byte inner MTU and 8192-entry rings, five interleaved 60-second samples per direction increased throughput by 16.9% forward and 18.0% reverse. Signed-off-by: Ralf Lici --- drivers/net/ovpn/io.h | 14 ++++++++++++++ drivers/net/ovpn/udp.c | 21 ++++++++++++++++++++- 2 files changed, 34 insertions(+), 1 deletion(-) diff --git a/drivers/net/ovpn/io.h b/drivers/net/ovpn/io.h index 1a94f0fda1d1..49180214fe08 100644 --- a/drivers/net/ovpn/io.h +++ b/drivers/net/ovpn/io.h @@ -10,6 +10,9 @@ #ifndef _NET_OVPN_OVPN_H_ #define _NET_OVPN_OVPN_H_ +#include +#include + /* DATA_V2 header size with AEAD encryption */ #define OVPN_HEAD_ROOM (OVPN_DATA_V2_OVERHEAD + \ max(sizeof(struct udphdr), sizeof(struct tcphdr)) +\ @@ -21,6 +24,17 @@ #define OVPN_KEEPALIVE_SIZE 16 extern const unsigned char ovpn_keepalive_message[OVPN_KEEPALIVE_SIZE]; +static inline void ovpn_skb_prefetchw(const struct sk_buff *skb) +{ + unsigned int offset; + + /* crypto overwrites data in place, so request write ownership of each + * linear cache line before the AEAD implementation reaches it + */ + for (offset = 0; offset < skb_headlen(skb); offset += L1_CACHE_BYTES) + prefetchw(skb->data + offset); +} + netdev_tx_t ovpn_net_xmit(struct sk_buff *skb, struct net_device *dev); void ovpn_recv(struct ovpn_peer *peer, struct sk_buff *skb); diff --git a/drivers/net/ovpn/udp.c b/drivers/net/ovpn/udp.c index dfb1aa10556d..31fae42e2990 100644 --- a/drivers/net/ovpn/udp.c +++ b/drivers/net/ovpn/udp.c @@ -32,6 +32,9 @@ /* like UDP and TCP frag-list GRO */ #define OVPN_UDP_GRO_CNT_MAX 64 +/* leave enough work between a cache hint and the record which consumes it */ +#define OVPN_UDP_GRO_PREFETCH_DISTANCE 2 + static bool ovpn_udp_gro_header(struct sk_buff *skb, u32 *header) { const unsigned int offset = skb_gro_offset(skb); @@ -171,12 +174,28 @@ static struct sk_buff *ovpn_udp_gro_detach(struct sk_buff *skb) static void ovpn_udp_recv(struct ovpn_peer *peer, struct sk_buff *skb) { - struct sk_buff *next; + struct sk_buff *next, *prefetch; + unsigned int i; skb->next = ovpn_udp_gro_detach(skb); + /* a frag-list GRO aggregate makes later ciphertext visible before the + * current record is decrypted, so we prime the first two records, then + * keep the cache hints the same distance ahead while draining the list + */ + prefetch = skb->next ? skb : NULL; + for (i = 0; i < OVPN_UDP_GRO_PREFETCH_DISTANCE && prefetch; i++) { + ovpn_skb_prefetchw(prefetch); + prefetch = prefetch->next; + } + skb_list_walk_safe(skb, skb, next) { + if (prefetch) { + ovpn_skb_prefetchw(prefetch); + prefetch = prefetch->next; + } + skb_mark_not_on_list(skb); /* keep the current reference alive for the next record before