From patchwork Tue Jul 13 20:57:53 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vasily Averin X-Patchwork-Id: 12375045 X-Patchwork-Delegate: kuba@kernel.org 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=-17.3 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER, INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 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 54C30C07E95 for ; Tue, 13 Jul 2021 20:57:58 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 3AAA660FED for ; Tue, 13 Jul 2021 20:57:58 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235490AbhGMVAq (ORCPT ); Tue, 13 Jul 2021 17:00:46 -0400 Received: from relay.sw.ru ([185.231.240.75]:56684 "EHLO relay.sw.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S235570AbhGMVAq (ORCPT ); Tue, 13 Jul 2021 17:00:46 -0400 DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=virtuozzo.com; s=relay; h=Content-Type:MIME-Version:Date:Message-ID:Subject :From; bh=wzD7p3eMscQSgryl/aNBA5sYqn+D84GYHaWrmA8jwMw=; b=yTpcUoq9HpBnTWTwb39 2vH3hZ8vDeWwIkIZPntvKC2/F9UNEKAl8bKI9/W66QzpWO7novbG/xioZUJkXoZj3VyAwAr7gcDre ZD53NHUeyKJ+CgHHlSwveDs9d0qZkHkltk+2KKsu+3fygQ5QT6Z0E1k6et7mNghTvXGrF6kza9A=; Received: from [10.93.0.56] by relay.sw.ru with esmtp (Exim 4.94.2) (envelope-from ) id 1m3PTW-003sYb-2W; Tue, 13 Jul 2021 23:57:54 +0300 From: Vasily Averin Subject: [PATCH NET v2 1/7] skbuff: introduce skb_expand_head() To: "David S. Miller" , Hideaki YOSHIFUJI , David Ahern , Jakub Kicinski , Eric Dumazet Cc: netdev@vger.kernel.org, Joerg Reuter , Ralf Baechle , linux-hams@vger.kernel.org, Alexei Starovoitov , Daniel Borkmann , Andrii Nakryiko , Martin KaFai Lau , Song Liu , Yonghong Song , KP Singh , bpf@vger.kernel.org, linux-kernel@vger.kernel.org References: <55c9e2ae-b060-baa2-460c-90eb3e9ded5c@virtuozzo.com> Message-ID: Date: Tue, 13 Jul 2021 23:57:53 +0300 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.11.0 MIME-Version: 1.0 In-Reply-To: Content-Language: en-US Precedence: bulk List-ID: X-Mailing-List: bpf@vger.kernel.org X-Patchwork-Delegate: kuba@kernel.org Like skb_realloc_headroom(), new helper increases headroom of specified skb. Unlike skb_realloc_headroom(), it does not allocate a new skb if possible; copies skb->sk on new skb when as needed and frees original skb in case of failures. This helps to simplify ip[6]_finish_output2() and a few other similar cases. Signed-off-by: Vasily Averin --- include/linux/skbuff.h | 1 + net/core/skbuff.c | 42 ++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 43 insertions(+) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index dbf820a..0003307 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1174,6 +1174,7 @@ static inline struct sk_buff *__pskb_copy(struct sk_buff *skb, int headroom, int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail, gfp_t gfp_mask); struct sk_buff *skb_realloc_headroom(struct sk_buff *skb, unsigned int headroom); +struct sk_buff *skb_expand_head(struct sk_buff *skb, unsigned int headroom); struct sk_buff *skb_copy_expand(const struct sk_buff *skb, int newheadroom, int newtailroom, gfp_t priority); int __must_check skb_to_sgvec_nomark(struct sk_buff *skb, struct scatterlist *sg, diff --git a/net/core/skbuff.c b/net/core/skbuff.c index bbc3b4b..a7997c2 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1769,6 +1769,48 @@ struct sk_buff *skb_realloc_headroom(struct sk_buff *skb, unsigned int headroom) EXPORT_SYMBOL(skb_realloc_headroom); /** + * skb_expand_head - reallocate header of &sk_buff + * @skb: buffer to reallocate + * @headroom: needed headroom + * + * Unlike skb_realloc_headroom, this one does not allocate a new skb + * if possible; copies skb->sk to new skb as needed + * and frees original skb in case of failures. + * + * It expect increased headroom and generates warning otherwise. + */ + +struct sk_buff *skb_expand_head(struct sk_buff *skb, unsigned int headroom) +{ + int delta = headroom - skb_headroom(skb); + + if (WARN_ONCE(delta <= 0, + "%s is expecting an increase in the headroom", __func__)) + return skb; + + /* pskb_expand_head() might crash, if skb is shared */ + if (skb_shared(skb)) { + struct sk_buff *nskb = skb_clone(skb, GFP_ATOMIC); + + if (likely(nskb)) { + if (skb->sk) + skb_set_owner_w(nskb, skb->sk); + consume_skb(skb); + } else { + kfree_skb(skb); + } + skb = nskb; + } + if (skb && + pskb_expand_head(skb, SKB_DATA_ALIGN(delta), 0, GFP_ATOMIC)) { + kfree_skb(skb); + skb = NULL; + } + return skb; +} +EXPORT_SYMBOL(skb_expand_head); + +/** * skb_copy_expand - copy and expand sk_buff * @skb: buffer to copy * @newheadroom: new free bytes at head From patchwork Tue Jul 13 20:58:03 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vasily Averin X-Patchwork-Id: 12375049 X-Patchwork-Delegate: kuba@kernel.org 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=-17.3 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER, INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 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 43F44C07E95 for ; Tue, 13 Jul 2021 20:58:08 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 308F9610A7 for ; Tue, 13 Jul 2021 20:58:08 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235846AbhGMVA5 (ORCPT ); Tue, 13 Jul 2021 17:00:57 -0400 Received: from relay.sw.ru ([185.231.240.75]:56738 "EHLO relay.sw.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S235772AbhGMVAz (ORCPT ); Tue, 13 Jul 2021 17:00:55 -0400 DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=virtuozzo.com; s=relay; h=Content-Type:MIME-Version:Date:Message-ID:Subject :From; bh=MiNg386dMSC6tinEQWI0fw04ONGjFF53u+lgA850DWw=; b=HIhsxuq3cGFjX7xKegI ps17JsNW8ZEwP5XWW7eoOFt/98YO3HrYyEVh3GVl9gxphBl/psXPGlov93nOHByH9gu8BoK5VeZbM zTM8CJsQPK0cqTho00D7C30vRK1reWRdqQdqccWvvp/NCJiokrv1qFzp9YTRY6LBNNTiNJ4vnfA=; Received: from [10.93.0.56] by relay.sw.ru with esmtp (Exim 4.94.2) (envelope-from ) id 1m3PTg-003sZT-40; Tue, 13 Jul 2021 23:58:04 +0300 From: Vasily Averin Subject: [PATCH NET v2 2/7] ipv6: use skb_expand_head in ip6_finish_output2 To: "David S. Miller" , Hideaki YOSHIFUJI , David Ahern , Jakub Kicinski , Eric Dumazet Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org References: <55c9e2ae-b060-baa2-460c-90eb3e9ded5c@virtuozzo.com> Message-ID: <99f76cb8-db3a-131b-18e3-f8cd3104c2d1@virtuozzo.com> Date: Tue, 13 Jul 2021 23:58:03 +0300 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.11.0 MIME-Version: 1.0 In-Reply-To: Content-Language: en-US Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org X-Patchwork-Delegate: kuba@kernel.org Unlike skb_realloc_headroom, new helper skb_expand_head does not allocate a new skb if possible. Additionally this patch replaces commonly used dereferencing with variables. --- NB: this patch depends on "ipv6: allocate enough headroom in ip6_finish_output2()" https://lkml.org/lkml/2021/7/12/732 Signed-off-by: Vasily Averin --- net/ipv6/ip6_output.c | 51 ++++++++++++++++----------------------------------- 1 file changed, 16 insertions(+), 35 deletions(-) diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 25144c7..6c4925e 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -60,46 +60,29 @@ static int ip6_finish_output2(struct net *net, struct sock *sk, struct sk_buff * { struct dst_entry *dst = skb_dst(skb); struct net_device *dev = dst->dev; + struct inet6_dev *idev = ip6_dst_idev(dst); unsigned int hh_len = LL_RESERVED_SPACE(dev); - int delta = hh_len - skb_headroom(skb); - const struct in6_addr *nexthop; + const struct in6_addr *daddr, *nexthop; + struct ipv6hdr *hdr; struct neighbour *neigh; int ret; /* Be paranoid, rather than too clever. */ - if (unlikely(delta > 0) && dev->header_ops) { - /* pskb_expand_head() might crash, if skb is shared */ - if (skb_shared(skb)) { - struct sk_buff *nskb = skb_clone(skb, GFP_ATOMIC); - - if (likely(nskb)) { - if (skb->sk) - skb_set_owner_w(nskb, skb->sk); - consume_skb(skb); - } else { - kfree_skb(skb); - } - skb = nskb; - } - if (skb && - pskb_expand_head(skb, SKB_DATA_ALIGN(delta), 0, GFP_ATOMIC)) { - kfree_skb(skb); - skb = NULL; - } + if (unlikely(hh_len > skb_headroom(skb)) && dev->header_ops) { + skb = skb_expand_head(skb, hh_len); if (!skb) { - IP6_INC_STATS(net, ip6_dst_idev(dst), IPSTATS_MIB_OUTDISCARDS); + IP6_INC_STATS(net, idev, IPSTATS_MIB_OUTDISCARDS); return -ENOMEM; } } - if (ipv6_addr_is_multicast(&ipv6_hdr(skb)->daddr)) { - struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb)); - + hdr = ipv6_hdr(skb); + daddr = &hdr->daddr; + if (ipv6_addr_is_multicast(daddr)) { if (!(dev->flags & IFF_LOOPBACK) && sk_mc_loop(sk) && ((mroute6_is_socket(net, skb) && !(IP6CB(skb)->flags & IP6SKB_FORWARDED)) || - ipv6_chk_mcast_addr(dev, &ipv6_hdr(skb)->daddr, - &ipv6_hdr(skb)->saddr))) { + ipv6_chk_mcast_addr(dev, daddr, &hdr->saddr))) { struct sk_buff *newskb = skb_clone(skb, GFP_ATOMIC); /* Do not check for IFF_ALLMULTI; multicast routing @@ -110,7 +93,7 @@ static int ip6_finish_output2(struct net *net, struct sock *sk, struct sk_buff * net, sk, newskb, NULL, newskb->dev, dev_loopback_xmit); - if (ipv6_hdr(skb)->hop_limit == 0) { + if (hdr->hop_limit == 0) { IP6_INC_STATS(net, idev, IPSTATS_MIB_OUTDISCARDS); kfree_skb(skb); @@ -119,9 +102,7 @@ static int ip6_finish_output2(struct net *net, struct sock *sk, struct sk_buff * } IP6_UPD_PO_STATS(net, idev, IPSTATS_MIB_OUTMCAST, skb->len); - - if (IPV6_ADDR_MC_SCOPE(&ipv6_hdr(skb)->daddr) <= - IPV6_ADDR_SCOPE_NODELOCAL && + if (IPV6_ADDR_MC_SCOPE(daddr) <= IPV6_ADDR_SCOPE_NODELOCAL && !(dev->flags & IFF_LOOPBACK)) { kfree_skb(skb); return 0; @@ -136,10 +117,10 @@ static int ip6_finish_output2(struct net *net, struct sock *sk, struct sk_buff * } rcu_read_lock_bh(); - nexthop = rt6_nexthop((struct rt6_info *)dst, &ipv6_hdr(skb)->daddr); - neigh = __ipv6_neigh_lookup_noref(dst->dev, nexthop); + nexthop = rt6_nexthop((struct rt6_info *)dst, daddr); + neigh = __ipv6_neigh_lookup_noref(dev, nexthop); if (unlikely(!neigh)) - neigh = __neigh_create(&nd_tbl, nexthop, dst->dev, false); + neigh = __neigh_create(&nd_tbl, nexthop, dev, false); if (!IS_ERR(neigh)) { sock_confirm_neigh(skb, neigh); ret = neigh_output(neigh, skb, false); @@ -148,7 +129,7 @@ static int ip6_finish_output2(struct net *net, struct sock *sk, struct sk_buff * } rcu_read_unlock_bh(); - IP6_INC_STATS(net, ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES); + IP6_INC_STATS(net, idev, IPSTATS_MIB_OUTNOROUTES); kfree_skb(skb); return -EINVAL; } From patchwork Tue Jul 13 20:58:10 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vasily Averin X-Patchwork-Id: 12375051 X-Patchwork-Delegate: kuba@kernel.org 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=-17.3 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER, INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 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 5072BC07E95 for ; Tue, 13 Jul 2021 20:58:16 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 3A98B6115B for ; Tue, 13 Jul 2021 20:58:16 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235933AbhGMVBF (ORCPT ); Tue, 13 Jul 2021 17:01:05 -0400 Received: from relay.sw.ru ([185.231.240.75]:56762 "EHLO relay.sw.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S235460AbhGMVBC (ORCPT ); Tue, 13 Jul 2021 17:01:02 -0400 DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=virtuozzo.com; s=relay; h=Content-Type:MIME-Version:Date:Message-ID:Subject :From; bh=WH01OAkO8I0mJuuKr6Adw0zwEAk80mXbraVHhNYTfAs=; b=xozvv1OKQ3oEt5enuSn 0FRmJlu/tqWwrErPB91NpVb7iDji/pB15CF8O3iPU4v1Hgif2wfjSJWZls8Qum2g63jSniKdkW5M7 5jBqWIhDt1Q1MG+zfGg1LZCkvvCbeVcICykAQB/JwchS4JDoBN+4jUIS/92HO8h/epIPnEvhtrw=; Received: from [10.93.0.56] by relay.sw.ru with esmtp (Exim 4.94.2) (envelope-from ) id 1m3PTn-003sZl-5C; Tue, 13 Jul 2021 23:58:11 +0300 From: Vasily Averin Subject: [PATCH NET v2 3/7] ipv6: use skb_expand_head in ip6_xmit To: "David S. Miller" , Hideaki YOSHIFUJI , David Ahern , Jakub Kicinski , Eric Dumazet Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org References: <55c9e2ae-b060-baa2-460c-90eb3e9ded5c@virtuozzo.com> Message-ID: <1cafc6e1-4e33-fa14-c726-7687eb0b59c6@virtuozzo.com> Date: Tue, 13 Jul 2021 23:58:10 +0300 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.11.0 MIME-Version: 1.0 In-Reply-To: Content-Language: en-US Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org X-Patchwork-Delegate: kuba@kernel.org Unlike skb_realloc_headroom, new helper skb_expand_head does not allocate a new skb if possible. Additionally this patch replaces commonly used dereferencing with variables. Signed-off-by: Vasily Averin --- net/ipv6/ip6_output.c | 27 +++++++++++---------------- 1 file changed, 11 insertions(+), 16 deletions(-) diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 6c4925e..90cd7b6 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -249,6 +249,8 @@ int ip6_xmit(const struct sock *sk, struct sk_buff *skb, struct flowi6 *fl6, const struct ipv6_pinfo *np = inet6_sk(sk); struct in6_addr *first_hop = &fl6->daddr; struct dst_entry *dst = skb_dst(skb); + struct net_device *dev = dst->dev; + struct inet6_dev *idev = ip6_dst_idev(dst); unsigned int head_room; struct ipv6hdr *hdr; u8 proto = fl6->flowi6_proto; @@ -256,22 +258,16 @@ int ip6_xmit(const struct sock *sk, struct sk_buff *skb, struct flowi6 *fl6, int hlimit = -1; u32 mtu; - head_room = sizeof(struct ipv6hdr) + LL_RESERVED_SPACE(dst->dev); + head_room = sizeof(struct ipv6hdr) + LL_RESERVED_SPACE(dev); if (opt) head_room += opt->opt_nflen + opt->opt_flen; - if (unlikely(skb_headroom(skb) < head_room)) { - struct sk_buff *skb2 = skb_realloc_headroom(skb, head_room); - if (!skb2) { - IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), - IPSTATS_MIB_OUTDISCARDS); - kfree_skb(skb); + if (unlikely(head_room > skb_headroom(skb))) { + skb = skb_expand_head(skb, head_room); + if (!skb) { + IP6_INC_STATS(net, idev, IPSTATS_MIB_OUTDISCARDS); return -ENOBUFS; } - if (skb->sk) - skb_set_owner_w(skb2, skb->sk); - consume_skb(skb); - skb = skb2; } if (opt) { @@ -313,8 +309,7 @@ int ip6_xmit(const struct sock *sk, struct sk_buff *skb, struct flowi6 *fl6, mtu = dst_mtu(dst); if ((skb->len <= mtu) || skb->ignore_df || skb_is_gso(skb)) { - IP6_UPD_PO_STATS(net, ip6_dst_idev(skb_dst(skb)), - IPSTATS_MIB_OUT, skb->len); + IP6_UPD_PO_STATS(net, idev, IPSTATS_MIB_OUT, skb->len); /* if egress device is enslaved to an L3 master device pass the * skb to its handler for processing @@ -327,17 +322,17 @@ int ip6_xmit(const struct sock *sk, struct sk_buff *skb, struct flowi6 *fl6, * we promote our socket to non const */ return NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, - net, (struct sock *)sk, skb, NULL, dst->dev, + net, (struct sock *)sk, skb, NULL, dev, dst_output); } - skb->dev = dst->dev; + skb->dev = dev; /* ipv6_local_error() does not require socket lock, * we promote our socket to non const */ ipv6_local_error((struct sock *)sk, EMSGSIZE, fl6, mtu); - IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGFAILS); + IP6_INC_STATS(net, idev, IPSTATS_MIB_FRAGFAILS); kfree_skb(skb); return -EMSGSIZE; } From patchwork Tue Jul 13 20:58:17 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vasily Averin X-Patchwork-Id: 12375053 X-Patchwork-Delegate: kuba@kernel.org 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=-17.3 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER, INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 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 87CF7C07E95 for ; Tue, 13 Jul 2021 20:58:24 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 7064361284 for ; Tue, 13 Jul 2021 20:58:24 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235960AbhGMVBN (ORCPT ); Tue, 13 Jul 2021 17:01:13 -0400 Received: from relay.sw.ru ([185.231.240.75]:56780 "EHLO relay.sw.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S235974AbhGMVBK (ORCPT ); Tue, 13 Jul 2021 17:01:10 -0400 DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=virtuozzo.com; s=relay; h=Content-Type:MIME-Version:Date:Message-ID:Subject :From; bh=LBHkRwPVEMlYau3k6VzHYQbsTeF/OcLEn5fTEBQXyLU=; b=HzdCc7DN85KT+fuvRTX 5V2AmQ9oI3KakPGeJ2qFfGVX+/+/Sl1fRJj6goIS75dY8Z3s95kjY9E6djDMDv3Wbg5Je+C/wr/Tx U1RqYbu/r+U7sGW+fkduQvGptFvAj4LjWCW0VUvnKQQHCIgQUcHknliY0Kivzhywx5b6Wh3eFrY=; Received: from [10.93.0.56] by relay.sw.ru with esmtp (Exim 4.94.2) (envelope-from ) id 1m3PTu-003sZz-VM; Tue, 13 Jul 2021 23:58:18 +0300 From: Vasily Averin Subject: [PATCH NET v2 4/7] ipv4: use skb_expand_head in ip_finish_output2 To: "David S. Miller" , Hideaki YOSHIFUJI , David Ahern , Jakub Kicinski , Eric Dumazet Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org References: <55c9e2ae-b060-baa2-460c-90eb3e9ded5c@virtuozzo.com> Message-ID: <9a21bf71-d224-a953-0648-47da136cb33b@virtuozzo.com> Date: Tue, 13 Jul 2021 23:58:17 +0300 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.11.0 MIME-Version: 1.0 In-Reply-To: Content-Language: en-US Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org X-Patchwork-Delegate: kuba@kernel.org Unlike skb_realloc_headroom, new helper skb_expand_head does not allocate a new skb if possible. Signed-off-by: Vasily Averin --- net/ipv4/ip_output.c | 13 ++----------- 1 file changed, 2 insertions(+), 11 deletions(-) diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index c3efc7d..5b2f6ea 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -198,19 +198,10 @@ static int ip_finish_output2(struct net *net, struct sock *sk, struct sk_buff *s } else if (rt->rt_type == RTN_BROADCAST) IP_UPD_PO_STATS(net, IPSTATS_MIB_OUTBCAST, skb->len); - /* Be paranoid, rather than too clever. */ if (unlikely(skb_headroom(skb) < hh_len && dev->header_ops)) { - struct sk_buff *skb2; - - skb2 = skb_realloc_headroom(skb, LL_RESERVED_SPACE(dev)); - if (!skb2) { - kfree_skb(skb); + skb = skb_expand_head(skb, hh_len); + if (!skb) return -ENOMEM; - } - if (skb->sk) - skb_set_owner_w(skb2, skb->sk); - consume_skb(skb); - skb = skb2; } if (lwtunnel_xmit_redirect(dst->lwtstate)) { From patchwork Tue Jul 13 20:58:28 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vasily Averin X-Patchwork-Id: 12375055 X-Patchwork-Delegate: kuba@kernel.org 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=-17.3 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER, INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 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 D9C64C07E95 for ; Tue, 13 Jul 2021 20:58:32 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id C405B60FED for ; Tue, 13 Jul 2021 20:58:32 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236085AbhGMVBW (ORCPT ); Tue, 13 Jul 2021 17:01:22 -0400 Received: from relay.sw.ru ([185.231.240.75]:56802 "EHLO relay.sw.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S234765AbhGMVBV (ORCPT ); Tue, 13 Jul 2021 17:01:21 -0400 DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=virtuozzo.com; s=relay; h=Content-Type:MIME-Version:Date:Message-ID:Subject :From; bh=DjCOdgpkoI0at8tuKwNz31O7tJL6hWslJd/79i+mUfg=; b=i4kDdCI+EgLRON8xYQm YUW/6jSjegOdiUvRv+h1vIAWIJU7kngB8PWw4fYWphEcz2jgwMCTkhJQ6SVnpUO3k4v8TP9AQXKVu 0cZOnAN8i62A2BKxVh78s17sYc7LugabnsOtcbEiDCfeaFOI2dQmHEqZ4Yu7GbJjJ66+8skPRWk=; Received: from [10.93.0.56] by relay.sw.ru with esmtp (Exim 4.94.2) (envelope-from ) id 1m3PU5-003saG-Nn; Tue, 13 Jul 2021 23:58:29 +0300 From: Vasily Averin Subject: [PATCH NET v2 5/7] vrf: use skb_expand_head in vrf_finish_output To: "David S. Miller" , Hideaki YOSHIFUJI , David Ahern , Jakub Kicinski , Eric Dumazet Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org References: <55c9e2ae-b060-baa2-460c-90eb3e9ded5c@virtuozzo.com> Message-ID: <85ba9cbe-2687-5af0-0b10-f06e35b44882@virtuozzo.com> Date: Tue, 13 Jul 2021 23:58:28 +0300 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.11.0 MIME-Version: 1.0 In-Reply-To: Content-Language: en-US Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org X-Patchwork-Delegate: kuba@kernel.org Unlike skb_realloc_headroom, new helper skb_expand_head does not allocate a new skb if possible. Signed-off-by: Vasily Averin --- drivers/net/vrf.c | 21 +++++++-------------- 1 file changed, 7 insertions(+), 14 deletions(-) diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index 28a6c4c..82e7696 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -857,30 +857,24 @@ static int vrf_finish_output(struct net *net, struct sock *sk, struct sk_buff *s unsigned int hh_len = LL_RESERVED_SPACE(dev); struct neighbour *neigh; bool is_v6gw = false; - int ret = -EINVAL; nf_reset_ct(skb); /* Be paranoid, rather than too clever. */ if (unlikely(skb_headroom(skb) < hh_len && dev->header_ops)) { - struct sk_buff *skb2; - - skb2 = skb_realloc_headroom(skb, LL_RESERVED_SPACE(dev)); - if (!skb2) { - ret = -ENOMEM; - goto err; + skb = skb_expand_head(skb, hh_len); + if (!skb) { + skb->dev->stats.tx_errors++; + return -ENOMEM; } - if (skb->sk) - skb_set_owner_w(skb2, skb->sk); - - consume_skb(skb); - skb = skb2; } rcu_read_lock_bh(); neigh = ip_neigh_for_gw(rt, skb, &is_v6gw); if (!IS_ERR(neigh)) { + int ret; + sock_confirm_neigh(skb, neigh); /* if crossing protocols, can not use the cached header */ ret = neigh_output(neigh, skb, is_v6gw); @@ -889,9 +883,8 @@ static int vrf_finish_output(struct net *net, struct sock *sk, struct sk_buff *s } rcu_read_unlock_bh(); -err: vrf_tx_error(skb->dev, skb); - return ret; + return -EINVAL; } static int vrf_output(struct net *net, struct sock *sk, struct sk_buff *skb) From patchwork Tue Jul 13 20:58:36 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vasily Averin X-Patchwork-Id: 12375057 X-Patchwork-Delegate: kuba@kernel.org 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=-17.3 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER, INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 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 5243EC07E95 for ; Tue, 13 Jul 2021 20:58:40 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 3AAE860FED for ; Tue, 13 Jul 2021 20:58:40 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236148AbhGMVB3 (ORCPT ); Tue, 13 Jul 2021 17:01:29 -0400 Received: from relay.sw.ru ([185.231.240.75]:56838 "EHLO relay.sw.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236038AbhGMVB2 (ORCPT ); Tue, 13 Jul 2021 17:01:28 -0400 DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=virtuozzo.com; s=relay; h=Content-Type:MIME-Version:Date:Message-ID:Subject :From; bh=pOmbvmoHCRf0LxILnkccI0/aUdAXXU2GWmHyziW8kQk=; b=RR7slBsKmKFzPKcpr5z 6igD9tUKJw1w5a+UPpq3hGO8pGpuKS4PT/rcVkxiDvhco8/e8CM70GDI8H093IeW2skEoxy4XDZDo WmUa+jeszHadQsCcZFsrht4zqQvNkszJ6uphMGeDGlrJldUTSMUl89y1WdzlhKV0sznLznUS5nA=; Received: from [10.93.0.56] by relay.sw.ru with esmtp (Exim 4.94.2) (envelope-from ) id 1m3PUC-003saU-MM; Tue, 13 Jul 2021 23:58:36 +0300 From: Vasily Averin Subject: [PATCH NET v2 6/7] ax25: use skb_expand_head To: "David S. Miller" , Hideaki YOSHIFUJI , David Ahern , Jakub Kicinski , Eric Dumazet Cc: netdev@vger.kernel.org, Joerg Reuter , Ralf Baechle , linux-hams@vger.kernel.org, linux-kernel@vger.kernel.org References: <55c9e2ae-b060-baa2-460c-90eb3e9ded5c@virtuozzo.com> Message-ID: <33cecb71-0582-308d-08b1-9445b4638ad7@virtuozzo.com> Date: Tue, 13 Jul 2021 23:58:36 +0300 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.11.0 MIME-Version: 1.0 In-Reply-To: Content-Language: en-US Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org X-Patchwork-Delegate: kuba@kernel.org Use skb_expand_head() in ax25_transmit_buffer and ax25_rt_build_path. Unlike skb_realloc_headroom, new helper does not allocate a new skb if possible. Signed-off-by: Vasily Averin --- v2: removed kfree_skb() in ax25_rt_build_path caller ax25_ip_xmit net/ax25/ax25_ip.c | 4 +--- net/ax25/ax25_out.c | 12 +++--------- net/ax25/ax25_route.c | 13 +++---------- 3 files changed, 7 insertions(+), 22 deletions(-) diff --git a/net/ax25/ax25_ip.c b/net/ax25/ax25_ip.c index e4f63dd..3624977 100644 --- a/net/ax25/ax25_ip.c +++ b/net/ax25/ax25_ip.c @@ -193,10 +193,8 @@ netdev_tx_t ax25_ip_xmit(struct sk_buff *skb) skb_pull(skb, AX25_KISS_HEADER_LEN); if (digipeat != NULL) { - if ((ourskb = ax25_rt_build_path(skb, src, dst, route->digipeat)) == NULL) { - kfree_skb(skb); + if ((ourskb = ax25_rt_build_path(skb, src, dst, route->digipeat)) == NULL) goto put; - } skb = ourskb; } diff --git a/net/ax25/ax25_out.c b/net/ax25/ax25_out.c index f53751b..af4a10e 100644 --- a/net/ax25/ax25_out.c +++ b/net/ax25/ax25_out.c @@ -336,18 +336,12 @@ void ax25_transmit_buffer(ax25_cb *ax25, struct sk_buff *skb, int type) headroom = ax25_addr_size(ax25->digipeat); - if (skb_headroom(skb) < headroom) { - if ((skbn = skb_realloc_headroom(skb, headroom)) == NULL) { + if (unlikely(skb_headroom(skb) < headroom)) { + skb = skb_expand_head(skb, headroom); + if (!skb) { printk(KERN_CRIT "AX.25: ax25_transmit_buffer - out of memory\n"); - kfree_skb(skb); return; } - - if (skb->sk != NULL) - skb_set_owner_w(skbn, skb->sk); - - consume_skb(skb); - skb = skbn; } ptr = skb_push(skb, headroom); diff --git a/net/ax25/ax25_route.c b/net/ax25/ax25_route.c index b40e0bc..d0b2e09 100644 --- a/net/ax25/ax25_route.c +++ b/net/ax25/ax25_route.c @@ -441,24 +441,17 @@ int ax25_rt_autobind(ax25_cb *ax25, ax25_address *addr) struct sk_buff *ax25_rt_build_path(struct sk_buff *skb, ax25_address *src, ax25_address *dest, ax25_digi *digi) { - struct sk_buff *skbn; unsigned char *bp; int len; len = digi->ndigi * AX25_ADDR_LEN; - if (skb_headroom(skb) < len) { - if ((skbn = skb_realloc_headroom(skb, len)) == NULL) { + if (unlikely(skb_headroom(skb) < len)) { + skb = skb_expand_head(skb, len); + if (!skb) { printk(KERN_CRIT "AX.25: ax25_dg_build_path - out of memory\n"); return NULL; } - - if (skb->sk != NULL) - skb_set_owner_w(skbn, skb->sk); - - consume_skb(skb); - - skb = skbn; } bp = skb_push(skb, len); From patchwork Tue Jul 13 20:58:42 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vasily Averin X-Patchwork-Id: 12375047 X-Patchwork-Delegate: kuba@kernel.org 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=-17.3 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER, INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 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 0682EC07E95 for ; Tue, 13 Jul 2021 20:58:47 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id E4F1A61152 for ; Tue, 13 Jul 2021 20:58:46 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236098AbhGMVBg (ORCPT ); Tue, 13 Jul 2021 17:01:36 -0400 Received: from relay.sw.ru ([185.231.240.75]:56872 "EHLO relay.sw.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S235060AbhGMVBf (ORCPT ); Tue, 13 Jul 2021 17:01:35 -0400 DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=virtuozzo.com; s=relay; h=Content-Type:MIME-Version:Date:Message-ID:Subject :From; bh=Zrmp38HLQSSBOIHCzrhOGN2I10/GGB7t9iYmdElr2i4=; b=oMb4Zx3XkPrKaEbUyM5 Wje5X6G4fTPCuW8lq2suLGXW4TgOCp3OBmHH00y2+c8S2v3SusR25Qn+9XuCa5+lRKj94uTuhRNP/ iXlyHbEjI3G1wn0S0pYAjM0i8kU7eYZurzXkFfp9sddH2FQI/V2+x06yB+T2hZ8RISMH1qTvwjw=; Received: from [10.93.0.56] by relay.sw.ru with esmtp (Exim 4.94.2) (envelope-from ) id 1m3PUJ-003san-IH; Tue, 13 Jul 2021 23:58:43 +0300 From: Vasily Averin Subject: [PATCH NET v2 7/7] bpf: use skb_expand_head in bpf_out_neigh_v4/6 To: "David S. Miller" , Hideaki YOSHIFUJI , David Ahern , Jakub Kicinski , Eric Dumazet Cc: netdev@vger.kernel.org, Alexei Starovoitov , Daniel Borkmann , Andrii Nakryiko , Martin KaFai Lau , Song Liu , Yonghong Song , KP Singh , bpf@vger.kernel.org, linux-kernel@vger.kernel.org References: <55c9e2ae-b060-baa2-460c-90eb3e9ded5c@virtuozzo.com> Message-ID: <47012e92-9978-cab1-68e4-9565ebe96994@virtuozzo.com> Date: Tue, 13 Jul 2021 23:58:42 +0300 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.11.0 MIME-Version: 1.0 In-Reply-To: Content-Language: en-US Precedence: bulk List-ID: X-Mailing-List: bpf@vger.kernel.org X-Patchwork-Delegate: kuba@kernel.org Unlike skb_realloc_headroom, new helper skb_expand_head does not allocate a new skb if possible. Additionally this patch replaces commonly used dereferencing with variables. Signed-off-by: Vasily Averin --- net/core/filter.c | 27 +++++---------------------- 1 file changed, 5 insertions(+), 22 deletions(-) diff --git a/net/core/filter.c b/net/core/filter.c index 65ab4e2..25a6950 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2179,17 +2179,9 @@ static int bpf_out_neigh_v6(struct net *net, struct sk_buff *skb, skb->tstamp = 0; if (unlikely(skb_headroom(skb) < hh_len && dev->header_ops)) { - struct sk_buff *skb2; - - skb2 = skb_realloc_headroom(skb, hh_len); - if (unlikely(!skb2)) { - kfree_skb(skb); + skb = skb_expand_head(skb, hh_len); + if (!skb) return -ENOMEM; - } - if (skb->sk) - skb_set_owner_w(skb2, skb->sk); - consume_skb(skb); - skb = skb2; } rcu_read_lock_bh(); @@ -2213,8 +2205,7 @@ static int bpf_out_neigh_v6(struct net *net, struct sk_buff *skb, } rcu_read_unlock_bh(); if (dst) - IP6_INC_STATS(dev_net(dst->dev), - ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES); + IP6_INC_STATS(net, ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES); out_drop: kfree_skb(skb); return -ENETDOWN; @@ -2286,17 +2277,9 @@ static int bpf_out_neigh_v4(struct net *net, struct sk_buff *skb, skb->tstamp = 0; if (unlikely(skb_headroom(skb) < hh_len && dev->header_ops)) { - struct sk_buff *skb2; - - skb2 = skb_realloc_headroom(skb, hh_len); - if (unlikely(!skb2)) { - kfree_skb(skb); + skb = skb_expand_head(skb, hh_len); + if (!skb) return -ENOMEM; - } - if (skb->sk) - skb_set_owner_w(skb2, skb->sk); - consume_skb(skb); - skb = skb2; } rcu_read_lock_bh();