From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 815373A5E7D; Sat, 17 Oct 2026 17:39:22 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1792258763; cv=none; b=cqD7jj4B+OPfO4z0im4mbBwxwMlzqExmSLApmYIlVr5TlWLsU6r4u+Wt1hvC25ZLqfTz4Je/TEoplM7L0DM8iOc7bayHS/z8WGBE3ZuShrKGL0pfGpixvvU854mQrvPUS5W+ZV14vqrTfblttDgMbe/WBEPVuNhxQi5PrHsGeVs= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1792258763; c=relaxed/simple; bh=EYwWx8l5R12k6io5V1LLSZHq3W1NnfdU6ZQRZ2PKtcw=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=oe/xn/kPG0FtIO5q4Uhap9hyHBMqYd5//rlWPvd+kmbB6a6Cu8KwD7D/Q/VtihMaN4FUY2Z9tAldSrVGLnW/sFI11jPTowUo7FGLmIXQyhKJm6fIfBk1GV/NBoR16FFKwhq+MzgA8dGTUpbQOuwlPoe7OXZubNXRdwSv2AZJJXI= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b=1bbxbJud; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b="1bbxbJud" Received: by smtp.kernel.org (Postfix) with ESMTPSA id D90C61F000C4; Sat, 17 Oct 2026 17:39:21 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linuxfoundation.org; s=korg; t=1792258762; bh=rTsRE0lPCwfRURR7amyDGfdorDoM+2KF4TGEnDZbVXY=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=1bbxbJudxZ2XQUrN9dWPt+qmFPXgxYwz5W8fXmbgqIyMUKA1Pl5lc2vPMdPMPugVR Fv3fNAtOzI+qOFX8uH+7XOyU/uvkLjXV884RIW1LAFL64AOQB4Wwx6K090R3oxRF/U gSMyqxu7RFvUu85ZE0rJtFd5X+o+AaBZ8SGjbc2g= From: Greg Kroah-Hartman To: stable@vger.kernel.org Cc: Greg Kroah-Hartman , patches@lists.linux.dev, Eric Dumazet , Joe Damato , Jakub Kicinski , Sasha Levin Subject: [PATCH 6.6 0991/1674] net: pull headers in qdisc_pkt_len_segs_init() Date: Sat, 17 Oct 2026 17:54:23 +0200 Message-ID: <20261017160456.556860899@linuxfoundation.org> X-Mailer: git-send-email 2.56.0 In-Reply-To: <20261017160431.119607952@linuxfoundation.org> References: <20261017160431.119607952@linuxfoundation.org> User-Agent: quilt/0.69 X-stable: review X-Patchwork-Hint: ignore Precedence: bulk X-Mailing-List: patches@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit 6.6-stable review patch. If anyone has any objections, please let me know. ------------------ From: Eric Dumazet [ Upstream commit 7fb4c19670110f052c04e1ec1d2b953b9f4f57e4 ] Most ndo_start_xmit() methods expects headers of gso packets to be already in skb->head. net/core/tso.c users are particularly at risk, because tso_build_hdr() does a memcpy(hdr, skb->data, hdr_len); qdisc_pkt_len_segs_init() already does a dissection of gso packets. Use pskb_may_pull() instead of skb_header_pointer() to make sure drivers do not have to reimplement this. Some malicious packets could be fed, detect them so that we can drop them sooner with a new SKB_DROP_REASON_SKB_BAD_GSO drop_reason. Fixes: e876f208af18 ("net: Add a software TSO helper API") Signed-off-by: Eric Dumazet Reviewed-by: Joe Damato Link: https://patch.msgid.link/20260403221540.3297753-3-edumazet@google.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin --- include/net/dropreason-core.h | 3 ++ net/core/dev.c | 81 +++++++++++++++++++++-------------- 2 files changed, 53 insertions(+), 31 deletions(-) diff --git a/include/net/dropreason-core.h b/include/net/dropreason-core.h index e88e785d221bb..16c140e00cc80 100644 --- a/include/net/dropreason-core.h +++ b/include/net/dropreason-core.h @@ -56,6 +56,7 @@ FN(UNHANDLED_PROTO) \ FN(SKB_CSUM) \ FN(SKB_GSO_SEG) \ + FN(SKB_BAD_GSO) \ FN(SKB_UCOPY_FAULT) \ FN(DEV_HDR) \ FN(DEV_READY) \ @@ -262,6 +263,8 @@ enum skb_drop_reason { SKB_DROP_REASON_SKB_CSUM, /** @SKB_DROP_REASON_SKB_GSO_SEG: gso segmentation error */ SKB_DROP_REASON_SKB_GSO_SEG, + /** @SKB_DROP_REASON_SKB_BAD_GSO: malicious gso packet. */ + SKB_DROP_REASON_SKB_BAD_GSO, /** * @SKB_DROP_REASON_SKB_UCOPY_FAULT: failed to copy data from user space, * e.g., via zerocopy_sg_from_iter() or skb_orphan_frags_rx() diff --git a/net/core/dev.c b/net/core/dev.c index 502b22c604693..045b815516010 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3821,49 +3821,62 @@ struct sk_buff *validate_xmit_skb_list(struct sk_buff *skb, struct net_device *d } EXPORT_SYMBOL_GPL(validate_xmit_skb_list); -static void qdisc_pkt_len_init(struct sk_buff *skb) +static enum skb_drop_reason qdisc_pkt_len_init(struct sk_buff *skb) { - const struct skb_shared_info *shinfo = skb_shinfo(skb); + struct skb_shared_info *shinfo = skb_shinfo(skb); + unsigned int hdr_len, tlen; + u16 gso_segs; qdisc_skb_cb(skb)->pkt_len = skb->len; + if (!shinfo->gso_size) + return SKB_NOT_DROPPED_YET; + + gso_segs = shinfo->gso_segs; /* To get more precise estimation of bytes sent on wire, * we add to pkt_len the headers size of all segments */ - if (shinfo->gso_size && skb_transport_header_was_set(skb)) { - u16 gso_segs = shinfo->gso_segs; - unsigned int hdr_len; - /* mac layer + network layer */ + /* mac layer + network layer */ + if (!skb->encapsulation) { + if (unlikely(!skb_transport_header_was_set(skb))) + return SKB_NOT_DROPPED_YET; hdr_len = skb_transport_offset(skb); + } else { + hdr_len = skb_inner_transport_offset(skb); + } + /* + transport layer */ + if (likely(shinfo->gso_type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6))) { + const struct tcphdr *th; - /* + transport layer */ - if (likely(shinfo->gso_type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6))) { - const struct tcphdr *th; - struct tcphdr _tcphdr; - - th = skb_header_pointer(skb, hdr_len, - sizeof(_tcphdr), &_tcphdr); - if (likely(th)) - hdr_len += __tcp_hdrlen(th); - } else if (shinfo->gso_type & SKB_GSO_UDP_L4) { - struct udphdr _udphdr; - - if (skb_header_pointer(skb, hdr_len, - sizeof(_udphdr), &_udphdr)) - hdr_len += sizeof(struct udphdr); - } + if (!pskb_may_pull(skb, hdr_len + sizeof(struct tcphdr))) + return SKB_DROP_REASON_SKB_BAD_GSO; - if (unlikely(shinfo->gso_type & SKB_GSO_DODGY)) { - int payload = skb->len - hdr_len; + th = (const struct tcphdr *)(skb->data + hdr_len); + tlen = __tcp_hdrlen(th); + if (tlen < sizeof(*th)) + return SKB_DROP_REASON_SKB_BAD_GSO; + hdr_len += tlen; + if (!pskb_may_pull(skb, hdr_len)) + return SKB_DROP_REASON_SKB_BAD_GSO; + } else if (shinfo->gso_type & SKB_GSO_UDP_L4) { + if (!pskb_may_pull(skb, hdr_len + sizeof(struct udphdr))) + return SKB_DROP_REASON_SKB_BAD_GSO; + hdr_len += sizeof(struct udphdr); + } - /* Malicious packet. */ - if (payload <= 0) - return; - gso_segs = DIV_ROUND_UP(payload, shinfo->gso_size); - } - qdisc_skb_cb(skb)->pkt_len += (gso_segs - 1) * hdr_len; + /* prior pskb_may_pull() might have changed skb->head. */ + shinfo = skb_shinfo(skb); + if (unlikely(shinfo->gso_type & SKB_GSO_DODGY)) { + int payload = skb->len - hdr_len; + + /* Malicious packet. */ + if (payload <= 0) + return SKB_DROP_REASON_SKB_BAD_GSO; + gso_segs = DIV_ROUND_UP(payload, shinfo->gso_size); } + qdisc_skb_cb(skb)->pkt_len += (gso_segs - 1) * hdr_len; + return SKB_NOT_DROPPED_YET; } static int dev_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *q, @@ -4407,6 +4420,7 @@ int __dev_queue_xmit(struct sk_buff *skb, struct net_device *sb_dev) struct Qdisc *q; int rc = -ENOMEM; bool again = false; + enum skb_drop_reason reason; skb_reset_mac_header(skb); skb_assert_len(skb); @@ -4414,6 +4428,12 @@ int __dev_queue_xmit(struct sk_buff *skb, struct net_device *sb_dev) if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_SCHED_TSTAMP)) __skb_tstamp_tx(skb, NULL, NULL, skb->sk, SCM_TSTAMP_SCHED); + reason = qdisc_pkt_len_init(skb); + if (unlikely(reason)) { + dev_core_stats_tx_dropped_inc(dev); + kfree_skb_reason(skb, reason); + return -EINVAL; + } /* Disable soft irqs for various locks below. Also * stops preemption for RCU. */ @@ -4421,7 +4441,6 @@ int __dev_queue_xmit(struct sk_buff *skb, struct net_device *sb_dev) skb_update_prio(skb); - qdisc_pkt_len_init(skb); tcx_set_ingress(skb, false); #ifdef CONFIG_NET_EGRESS if (static_branch_unlikely(&egress_needed_key)) { -- 2.53.0