From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-qk1-f170.google.com (mail-qk1-f170.google.com [209.85.222.170]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 79557335BA for ; Sun, 31 May 2026 12:33:04 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.222.170 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1780230785; cv=none; b=lXAKuacUk8oKYEb8ctttDQH1Yle8NkWrKmXh+aTyustXx8Kw6E5PH6WXZoYAyWXP1TacmZbcUUBCeie8nDidAsyEVef+z4jBMjtj8g33c7ESH+kiPuhyNWx5SQC2KcZb3xSYKgkpBYMegZ9C0Lf7NtiNyPUgoSbyuyh6tBSf+Nc= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1780230785; c=relaxed/simple; bh=niiFf0bi+TOBAfz2k1dJeC/8jZl57nPOEYNIfxVFaPc=; h=From:To:Cc:Subject:Date:Message-Id:MIME-Version:Content-Type; b=fFzo9cUZhez0Fj1OnPp5yttqGm/OiCLTzsHT2sV194F6+iPfJpVpiTDSGAKjOSan8VF2vwj4VnA6tS//0MTIyZ1cncrFxOZ8DInvX1r18IlIlO+O38yw72kgnWjncVOWw6Z2xqRjBuzgALdKUlhpa9fPnb5TzVsrRbHg/1XxeoE= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=mojatatu.com; spf=none smtp.mailfrom=mojatatu.com; dkim=pass (1024-bit key) header.d=mojatatu.com header.i=@mojatatu.com header.b=BXVcTMYr; arc=none smtp.client-ip=209.85.222.170 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=mojatatu.com Authentication-Results: smtp.subspace.kernel.org; spf=none smtp.mailfrom=mojatatu.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=mojatatu.com header.i=@mojatatu.com header.b="BXVcTMYr" Received: by mail-qk1-f170.google.com with SMTP id af79cd13be357-9153e2708aeso156458585a.0 for ; Sun, 31 May 2026 05:33:04 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=mojatatu.com; s=google; t=1780230783; x=1780835583; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:message-id:date:subject:cc :to:from:from:to:cc:subject:date:message-id:reply-to; bh=Tnbx1FygkSpUfgjfi784cy+83ZrekiRaaK9vHrANUOo=; b=BXVcTMYrMLMphiFRlQ3tRE5d4qjQEgJDvaAkPY0GGGVNguZNW5KXeKpWUirIhHwZdS T5+KHwPojobkqgRSWMEmk96H2bGD0XBXh+86OEf8TZbcpSGrUONFpFjkVH23/lcEaXD8 QlJCzVO2fN7wVnxUInp3hcncgOUqvF5aa/Cf0= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1780230783; x=1780835583; h=content-transfer-encoding:mime-version:message-id:date:subject:cc :to:from:x-gm-gg:x-gm-message-state:from:to:cc:subject:date :message-id:reply-to; bh=Tnbx1FygkSpUfgjfi784cy+83ZrekiRaaK9vHrANUOo=; b=b6EH3LdfW+H/pk7VG9AwVIpZHpbYINlJmUA7ZD58Hui6wW1fbwa4+toPKRYMfs+xxT SaatJZl2iZzHt1/C2/p3mWRwjKWQkX0ZPMRRQ2rJka1nWuj2R9iq8ZX66F/GCD2gP0lE V1ODtlegXQMHvSSsUsewZNFEQxixkNZXA1hq5jwdw2tDRDGd4Z1MNnt1qB3vP/0KTz41 GkfQXSa5F1sHq7K37SgpvUCAuN6vDJvVKWNQeVta9h06BBP0VAX6uBFBK/m3R+7X+5sA +xIGL+j1yjXFKN/gfPwfxK62+MIGFhM0c0WoJnSRFu7KypJbSxVY26ys2Jq0W4KrWihe omFQ== X-Forwarded-Encrypted: i=1; AFNElJ/UMs+ENyoCw0G/KCJr83ByPlZXx0PBjmfStgQHk/tLQnj8Byoiw+2JJ3WSSe2ncZamLlykLkybFOgCjwk=@vger.kernel.org X-Gm-Message-State: AOJu0Yz1C0fe2r7DK7i1M3vcC9m/lVrEtekgAeXCmGE9xiloVBRy47NN kpyD4NT24S/Y4EsT8+ZeCFhYLz1hm8X0O7kaju5VIvEbnOov01hMxJ9OakHiqEu06Q== X-Gm-Gg: Acq92OHNEcEW6coPjbuw/37sCr5rHlNy+BBFv7pqBxlfpIFiUOP0Zp1m/YaiXpvbhNQ b6SqzpO7MBpq16z5W7v6sM4xGaEdmg92YS1u74meWpfM/5/Xf5txdxhctX/unyVHAF1atEf0UEo 69pybGyxoKeILEB9qvCr3fbLWt+5WDkDiM2AW++l7dC9Fmv4ADNVVGzWggjPMGF/OsdHmykGKz6 xGrOzx59y5M0A1gC5zCjGBNjGLtC9KXRusXpAH/OtHR5Zdz4RE34z+jwBQlxn3MS1YD87WiRszl AzOOXeNMvwaFGaLPOGbjQz+lddeudSMdu/d8HpIKXytdGGkynduVFL3RvIHXJMgab9GpoG9ppOB tbap8DzVGlJatHkhrztB6+v76RO/gQLLclqYAzA7gLWb9VnDUlizVWhdetm1ORF/3AxRq2sUi+8 /A7Pxzh+7b8ZXrrqsSRUuNt2ghnlM= X-Received: by 2002:a05:620a:708e:b0:8ea:d123:5636 with SMTP id af79cd13be357-9153da34264mr1049046785a.31.1780230783498; Sun, 31 May 2026 05:33:03 -0700 (PDT) Received: from majuu.waya ([184.144.29.222]) by smtp.gmail.com with ESMTPSA id af79cd13be357-915326016c2sm728127285a.25.2026.05.31.05.33.01 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Sun, 31 May 2026 05:33:02 -0700 (PDT) From: Jamal Hadi Salim To: netdev@vger.kernel.org Cc: davem@davemloft.net, edumazet@google.com, kuba@kernel.org, pabeni@redhat.com, horms@kernel.org, jiri@resnulli.us, victor@mojatatu.com, david.laight.linux@gmail.com, yimingqian591@gmail.com, keenanat2000@gmail.com, 2045gemini@gmail.com, rollkingzzc@gmail.com, toke@redhat.com, dcaratti@redhat.com, security@kernel.org, linux-kernel@vger.kernel.org, stable@kernel.org, Rajat Gupta , Jamal Hadi Salim Subject: [PATCH net v5 1/1] net/sched: fix pedit partial COW leading to page cache corruption Date: Sun, 31 May 2026 08:32:21 -0400 Message-Id: <20260531123221.48732-1-jhs@mojatatu.com> X-Mailer: git-send-email 2.34.1 Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit From: Rajat Gupta tcf_pedit_act() computes the COW range for skb_ensure_writable() once before the key loop using tcfp_off_max_hint, but the hint does not account for the runtime header offset added by typed keys. This can leave part of the write region un-COW'd. Fix by moving skb_ensure_writable() inside the per-key loop where the actual write offset is known, and add overflow checking on the offset arithmetic. For negative offsets (e.g. Ethernet header edits at ingress), use skb_cow() to COW the headroom instead. Guard offset_valid() against INT_MIN, where negation is undefined. Fixes: 8b796475fd78 ("net/sched: act_pedit: really ensure the skb is writable") Reported-by: Yiming Qian Reported-by: Keenan Dong Reported-by: Han Guidong <2045gemini@gmail.com> Reported-by: Zhang Cen Reviewed-by: Han Guidong <2045gemini@gmail.com> Tested-by: Han Guidong <2045gemini@gmail.com> Reviewed-by: Davide Caratti Tested-by: Davide Caratti Reviewed-by: Toke Høiland-Jørgensen Tested-by: Toke Høiland-Jørgensen Reviewed-by: Victor Nogueira Tested-by: Victor Nogueira Acked-by: Jamal Hadi Salim Signed-off-by: Rajat Gupta --- v4->v5: 1) Dead code removal obsoleted after removing tcfp_off_max_hint as identified by sashiko-claude[1]. Remove dead code updating "cur". 2) Addition of hoffset at tkey->at promotes hoffset to unsigned int, wraps on overflow. Claim made by both sashiko-claude[1] and sashiko-gemini[2] I think this is far fetched but possible. Cast tkey->at to int. 3) Signedness mismatch in min() macro may potentially cause build issues. Claim made by both sashiko-claude[1] and sashiko-gemini[2]. Go back to min_t() for now. David L. can make a better proposal later.. [1]https://netdev-ai.bots.linux.dev/sashiko/#/patchset/20260530080643.1345521-1-jhs%40mojatatu.com [2]https://sashiko.dev/#/patchset/20260530080643.1345521-1-jhs%40mojatatu.com --- include/net/tc_act/tc_pedit.h | 1 - net/sched/act_pedit.c | 77 +++++++++++++++++++---------------- 2 files changed, 41 insertions(+), 37 deletions(-) diff --git a/include/net/tc_act/tc_pedit.h b/include/net/tc_act/tc_pedit.h index f58ee15cd858..cb7b82f2cbc7 100644 --- a/include/net/tc_act/tc_pedit.h +++ b/include/net/tc_act/tc_pedit.h @@ -15,7 +15,6 @@ struct tcf_pedit_parms { struct tc_pedit_key *tcfp_keys; struct tcf_pedit_key_ex *tcfp_keys_ex; int action; - u32 tcfp_off_max_hint; unsigned char tcfp_nkeys; unsigned char tcfp_flags; struct rcu_head rcu; diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c index bc20f08a2789..bd3b1da3cd63 100644 --- a/net/sched/act_pedit.c +++ b/net/sched/act_pedit.c @@ -16,6 +16,8 @@ #include #include #include +#include +#include #include #include #include @@ -242,7 +244,6 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla, goto out_free_ex; } - nparms->tcfp_off_max_hint = 0; nparms->tcfp_flags = parm->flags; nparms->tcfp_nkeys = parm->nkeys; @@ -268,14 +269,6 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla, BITS_PER_TYPE(int) - 1, nparms->tcfp_keys[i].shift); - /* The AT option can read a single byte, we can bound the actual - * value with uchar max. - */ - cur += (0xff & offmask) >> nparms->tcfp_keys[i].shift; - - /* Each key touches 4 bytes starting from the computed offset */ - nparms->tcfp_off_max_hint = - max(nparms->tcfp_off_max_hint, cur + 4); } p = to_pedit(*a); @@ -318,15 +311,12 @@ static void tcf_pedit_cleanup(struct tc_action *a) call_rcu(&parms->rcu, tcf_pedit_cleanup_rcu); } -static bool offset_valid(struct sk_buff *skb, int offset) +static bool offset_valid(struct sk_buff *skb, int offset, int len) { - if (offset > 0 && offset > skb->len) - return false; - - if (offset < 0 && -offset > skb_headroom(skb)) + if (offset < -(int)skb_headroom(skb)) return false; - return true; + return offset <= (int)skb->len - len; } static int pedit_l4_skb_offset(struct sk_buff *skb, int *hoffset, const int header_type) @@ -393,18 +383,10 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb, struct tcf_pedit_key_ex *tkey_ex; struct tcf_pedit_parms *parms; struct tc_pedit_key *tkey; - u32 max_offset; int i; parms = rcu_dereference_bh(p->parms); - max_offset = (skb_transport_header_was_set(skb) ? - skb_transport_offset(skb) : - skb_network_offset(skb)) + - parms->tcfp_off_max_hint; - if (skb_ensure_writable(skb, min(skb->len, max_offset))) - goto done; - tcf_lastuse_update(&p->tcf_tm); tcf_action_update_bstats(&p->common, skb); @@ -412,10 +394,11 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb, tkey_ex = parms->tcfp_keys_ex; for (i = parms->tcfp_nkeys; i > 0; i--, tkey++) { + int write_offset, write_len; int offset = tkey->off; int hoffset = 0; - u32 *ptr, hdata; - u32 val; + u32 cur_val, val; + u32 *ptr; int rc; if (tkey_ex) { @@ -433,13 +416,15 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb, if (tkey->offmask) { u8 *d, _d; + int at_offset; - if (!offset_valid(skb, hoffset + tkey->at)) { + if (check_add_overflow(hoffset, (int)tkey->at, &at_offset) || + !offset_valid(skb, at_offset, sizeof(_d))) { pr_info_ratelimited("tc action pedit 'at' offset %d out of bounds\n", hoffset + tkey->at); goto bad; } - d = skb_header_pointer(skb, hoffset + tkey->at, + d = skb_header_pointer(skb, at_offset, sizeof(_d), &_d); if (!d) goto bad; @@ -451,31 +436,51 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb, } } - if (!offset_valid(skb, hoffset + offset)) { - pr_info_ratelimited("tc action pedit offset %d out of bounds\n", hoffset + offset); + if (check_add_overflow(hoffset, offset, &write_offset)) { + pr_info_ratelimited("tc action pedit offset overflow\n"); goto bad; } - ptr = skb_header_pointer(skb, hoffset + offset, - sizeof(hdata), &hdata); - if (!ptr) + if (!offset_valid(skb, write_offset, sizeof(*ptr))) { + pr_info_ratelimited("tc action pedit offset %d out of bounds\n", + write_offset); goto bad; + } + + if (write_offset < 0) { + if (skb_cow(skb, -write_offset)) + goto bad; + if (write_offset + (int)sizeof(*ptr) > 0) { + if (skb_ensure_writable(skb, + min_t(int, skb->len, + write_offset + (int)sizeof(*ptr)))) + goto bad; + } + } else { + if (check_add_overflow(write_offset, (int)sizeof(*ptr), + &write_len)) + goto bad; + if (skb_ensure_writable(skb, min_t(int, skb->len, + write_len))) + goto bad; + } + + ptr = (u32 *)(skb->data + write_offset); + cur_val = get_unaligned(ptr); /* just do it, baby */ switch (cmd) { case TCA_PEDIT_KEY_EX_CMD_SET: val = tkey->val; break; case TCA_PEDIT_KEY_EX_CMD_ADD: - val = (*ptr + tkey->val) & ~tkey->mask; + val = (cur_val + tkey->val) & ~tkey->mask; break; default: pr_info_ratelimited("tc action pedit bad command (%d)\n", cmd); goto bad; } - *ptr = ((*ptr & tkey->mask) ^ val); - if (ptr == &hdata) - skb_store_bits(skb, hoffset + offset, ptr, 4); + put_unaligned((cur_val & tkey->mask) ^ val, ptr); } goto done; -- 2.34.1