From nobody Wed May 22 02:33:18 2024 Received: from mx0a-0031df01.pphosted.com (mx0a-0031df01.pphosted.com [205.220.168.131]) (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 DABBF5660; Thu, 18 Apr 2024 00:43:34 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=205.220.168.131 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1713401016; cv=none; b=lygy+hhDZ/hyPracLt+vuRoviVPWYD7mwDPJDmpHeJThwV4VrgC5b6gVRAtBLvVv9v5N/KnEj6jb2X5pAHwXNs+rliEOTKXXQuOBThuZlrkYtIgzNFoF9bZFdmZtEWpy/bVfg7Uf3Ww+nu933motmyPSuvYlVm78yE81VNqwslk= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1713401016; c=relaxed/simple; bh=JDegKX+4p7svYdktxGs6mSFTPoz9AITMli9g33YO4a4=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=TukJ0rnruu1vEuwdFlyCpyqP3c57Ndnl9/2/gctqTJE+H+G6q69m6xQooHuHIWrAqfyKUklwt1stTzej+3hzdKJcsMgh4gKSd9rgMLC6Xh6HEjoohOmZPCJg5tTCwAwhRZoczBwqLKHqRGNp9qkAJUay7YchF/cIvg/AFIj0TYM= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=quicinc.com; spf=pass smtp.mailfrom=qualcomm.com; dkim=pass (2048-bit key) header.d=quicinc.com header.i=@quicinc.com header.b=Sk1EbVVG; arc=none smtp.client-ip=205.220.168.131 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=quicinc.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=qualcomm.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=quicinc.com header.i=@quicinc.com header.b="Sk1EbVVG" Received: from pps.filterd (m0279864.ppops.net [127.0.0.1]) by mx0a-0031df01.pphosted.com (8.17.1.24/8.17.1.24) with ESMTP id 43HNqQ88003134; Thu, 18 Apr 2024 00:43:12 GMT DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=quicinc.com; h= from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; s=qcppdkim1; bh=ks+KT4K 5fNaV+wmbebmRWsM6NFf4ZLUoDmy/5G+3PfY=; b=Sk1EbVVGaJ4VCC3Bd461ENV OMp2ud4grXCH4QYxEnTuk2sFlDO2GAMsIhJAcf9eqn0AEEEi5viEysbdhhxlaAGy n4tpNKGFHURWQGD83DawcV0357xmaqNaS/TwTaV//0gFoEShAJtByA7O8M7lliUL 4xZyZq/4wM9f2xCR1TXtWoVROHysohHBGdfPOKB7H4qfkyBlt0bhDfxMsuvwuQG/ AOS0+zLu4LnfVzznDd4Y7sJ50RPd1T/SNlbtNMTKrPlI83AABztsxE2mylq/8QRM 3oliV7RPxHuxbZAVgzSVJU8J0RGAo8h+IU/R38kgBus3BmwhNhpXmd87HJUOmXg= = Received: from nalasppmta01.qualcomm.com (Global_NAT1.qualcomm.com [129.46.96.20]) by mx0a-0031df01.pphosted.com (PPS) with ESMTPS id 3xjm0s0nkp-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Thu, 18 Apr 2024 00:43:12 +0000 (GMT) Received: from pps.filterd (NALASPPMTA01.qualcomm.com [127.0.0.1]) by NALASPPMTA01.qualcomm.com (8.17.1.5/8.17.1.5) with ESMTP id 43I0fCTK024654; Thu, 18 Apr 2024 00:43:09 GMT Received: from pps.reinject (localhost [127.0.0.1]) by NALASPPMTA01.qualcomm.com (PPS) with ESMTPS id 3xhq5fehg4-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Thu, 18 Apr 2024 00:43:09 +0000 Received: from NALASPPMTA01.qualcomm.com (NALASPPMTA01.qualcomm.com [127.0.0.1]) by pps.reinject (8.17.1.5/8.17.1.5) with ESMTP id 43I0h9w3028035; Thu, 18 Apr 2024 00:43:09 GMT Received: from hu-devc-lv-u20-a-new.qualcomm.com (hu-abchauha-lv.qualcomm.com [10.81.25.35]) by NALASPPMTA01.qualcomm.com (PPS) with ESMTPS id 43I0h9p2028025 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Thu, 18 Apr 2024 00:43:09 +0000 Received: by hu-devc-lv-u20-a-new.qualcomm.com (Postfix, from userid 214165) id B883C21C16; Wed, 17 Apr 2024 17:43:08 -0700 (PDT) From: Abhishek Chauhan To: "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , netdev@vger.kernel.org, linux-kernel@vger.kernel.org, Andrew Halaney , Willem de Bruijn , Martin KaFai Lau , Martin KaFai Lau , Daniel Borkmann , bpf Cc: kernel@quicinc.com Subject: [RFC PATCH bpf-next v4 1/2] net: Rename mono_delivery_time to tstamp_type for scalabilty Date: Wed, 17 Apr 2024 17:43:07 -0700 Message-Id: <20240418004308.1009262-2-quic_abchauha@quicinc.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20240418004308.1009262-1-quic_abchauha@quicinc.com> References: <20240418004308.1009262-1-quic_abchauha@quicinc.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-QCInternal: smtphost X-QCInternal: smtphost X-Proofpoint-Virus-Version: vendor=nai engine=6200 definitions=5800 signatures=585085 X-Proofpoint-Virus-Version: vendor=nai engine=6200 definitions=5800 signatures=585085 X-Proofpoint-GUID: DWS6YolHOEPsN7hly-08AOGCA6lJILzQ X-Proofpoint-ORIG-GUID: DWS6YolHOEPsN7hly-08AOGCA6lJILzQ X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.272,Aquarius:18.0.1011,Hydra:6.0.619,FMLib:17.11.176.26 definitions=2024-04-17_20,2024-04-17_01,2023-05-22_02 X-Proofpoint-Spam-Details: rule=outbound_notspam policy=outbound score=0 lowpriorityscore=0 bulkscore=0 adultscore=0 impostorscore=0 mlxscore=0 mlxlogscore=999 spamscore=0 priorityscore=1501 phishscore=0 malwarescore=0 clxscore=1015 suspectscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.19.0-2404010003 definitions=main-2404180003 Content-Type: text/plain; charset="utf-8" mono_delivery_time was added to check if skb->tstamp has delivery time in mono clock base (i.e. EDT) otherwise skb->tstamp has timestamp in ingress and delivery_time at egress. Renaming the bitfield from mono_delivery_time to tstamp_type is for extensibilty for other timestamps such as userspace timestamp (i.e. SO_TXTIME) set via sock opts. As we are renaming the mono_delivery_time to tstamp_type, it makes sense to start assigning tstamp_type based on enum defined in this commit. Earlier we used bool arg flag to check if the tstamp is mono in function skb_set_delivery_time, Now the signature of the functions accepts tstamp_type to distinguish between mono and real time. In future tstamp_type:1 can be extended to support userspace timestamp by increasing the bitfield. Link: https://lore.kernel.org/netdev/bc037db4-58bb-4861-ac31-a361a93841d3@l= inux.dev/ Signed-off-by: Abhishek Chauhan --- Changes since v3 - Fixed inconsistent capitalization in skbuff.h - remove reference to MONO_DELIVERY_TIME_MASK in skbuff.h and point it to skb_tstamp_type now. - Explicitely setting SKB_CLOCK_MONO if valid transmit_time ip_send_unicast_reply=20 - Keeping skb_tstamp inline with skb_clear_tstamp.=20 - skb_set_delivery_time checks if timstamp is 0 and=20 sets the tstamp_type to SKB_CLOCK_REAL. - Above comments are given by Willem=20 - Found out that skbuff.h has access to uapi/linux/time.h So now instead of using CLOCK_REAL/CLOCK_MONO=20 i am checking actual clockid_t directly to set tstamp_type=20 example:- CLOCK_REALTIME/CLOCK_MONOTONIC=20 - Compilation error fixed in=20 net/ieee802154/6lowpan/reassembly.c Changes since v2 - Minor changes to commit subject Changes since v1 - Squashed the two commits into one as mentioned by Willem. - Introduced switch in skb_set_delivery_time. - Renamed and removed directionality aspects w.r.t tstamp_type=20 as mentioned by Willem. include/linux/skbuff.h | 44 +++++++++++++++++----- include/net/inet_frag.h | 4 +- net/bridge/netfilter/nf_conntrack_bridge.c | 6 +-- net/core/dev.c | 2 +- net/core/filter.c | 10 ++--- net/ieee802154/6lowpan/reassembly.c | 2 +- net/ipv4/inet_fragment.c | 2 +- net/ipv4/ip_fragment.c | 2 +- net/ipv4/ip_output.c | 9 +++-- net/ipv4/tcp_output.c | 14 +++---- net/ipv6/ip6_output.c | 6 +-- net/ipv6/netfilter.c | 6 +-- net/ipv6/netfilter/nf_conntrack_reasm.c | 2 +- net/ipv6/reassembly.c | 2 +- net/ipv6/tcp_ipv6.c | 2 +- net/sched/act_bpf.c | 4 +- net/sched/cls_bpf.c | 4 +- 17 files changed, 73 insertions(+), 48 deletions(-) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index aded4949ea79..2e7811c7e4db 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -702,6 +702,17 @@ typedef unsigned int sk_buff_data_t; typedef unsigned char *sk_buff_data_t; #endif =20 +/** + * tstamp_type:1 can take 2 values each + * represented by time base in skb + * 0x0 =3D> real timestamp_type + * 0x1 =3D> mono timestamp_type + */ +enum skb_tstamp_type { + SKB_CLOCK_REAL, /* Time base is skb is REALTIME */ + SKB_CLOCK_MONO, /* Time base is skb is MONOTONIC */ +}; + /** * DOC: Basic sk_buff geometry * @@ -819,7 +830,7 @@ typedef unsigned char *sk_buff_data_t; * @dst_pending_confirm: need to confirm neighbour * @decrypted: Decrypted SKB * @slow_gro: state present at GRO time, slower prepare step required - * @mono_delivery_time: When set, skb->tstamp has the + * @tstamp_type: When set, skb->tstamp has the * delivery_time in mono clock base (i.e. EDT). Otherwise, the * skb->tstamp has the (rcv) timestamp at ingress and * delivery_time at egress. @@ -950,7 +961,7 @@ struct sk_buff { /* private: */ __u8 __mono_tc_offset[0]; /* public: */ - __u8 mono_delivery_time:1; /* See SKB_MONO_DELIVERY_TIME_MASK */ + __u8 tstamp_type:1; /* See SKB_CLOCK_*_MASK */ #ifdef CONFIG_NET_XGRESS __u8 tc_at_ingress:1; /* See TC_AT_INGRESS_MASK */ __u8 tc_skip_classify:1; @@ -4170,7 +4181,7 @@ static inline void skb_get_new_timestampns(const stru= ct sk_buff *skb, static inline void __net_timestamp(struct sk_buff *skb) { skb->tstamp =3D ktime_get_real(); - skb->mono_delivery_time =3D 0; + skb->tstamp_type =3D SKB_CLOCK_REAL; } =20 static inline ktime_t net_timedelta(ktime_t t) @@ -4179,10 +4190,23 @@ static inline ktime_t net_timedelta(ktime_t t) } =20 static inline void skb_set_delivery_time(struct sk_buff *skb, ktime_t kt, - bool mono) + u8 tstamp_type) { skb->tstamp =3D kt; - skb->mono_delivery_time =3D kt && mono; + + if (!kt) { + skb->tstamp_type =3D SKB_CLOCK_REAL; + return; + } + + switch (tstamp_type) { + case CLOCK_REALTIME: + skb->tstamp_type =3D SKB_CLOCK_REAL; + break; + case CLOCK_MONOTONIC: + skb->tstamp_type =3D SKB_CLOCK_MONO; + break; + } } =20 DECLARE_STATIC_KEY_FALSE(netstamp_needed_key); @@ -4192,8 +4216,8 @@ DECLARE_STATIC_KEY_FALSE(netstamp_needed_key); */ static inline void skb_clear_delivery_time(struct sk_buff *skb) { - if (skb->mono_delivery_time) { - skb->mono_delivery_time =3D 0; + if (skb->tstamp_type) { + skb->tstamp_type =3D SKB_CLOCK_REAL; if (static_branch_unlikely(&netstamp_needed_key)) skb->tstamp =3D ktime_get_real(); else @@ -4203,7 +4227,7 @@ static inline void skb_clear_delivery_time(struct sk_= buff *skb) =20 static inline void skb_clear_tstamp(struct sk_buff *skb) { - if (skb->mono_delivery_time) + if (skb->tstamp_type) return; =20 skb->tstamp =3D 0; @@ -4211,7 +4235,7 @@ static inline void skb_clear_tstamp(struct sk_buff *s= kb) =20 static inline ktime_t skb_tstamp(const struct sk_buff *skb) { - if (skb->mono_delivery_time) + if (skb->tstamp_type) return 0; =20 return skb->tstamp; @@ -4219,7 +4243,7 @@ static inline ktime_t skb_tstamp(const struct sk_buff= *skb) =20 static inline ktime_t skb_tstamp_cond(const struct sk_buff *skb, bool cond) { - if (!skb->mono_delivery_time && skb->tstamp) + if (skb->tstamp_type !=3D SKB_CLOCK_MONO && skb->tstamp) return skb->tstamp; =20 if (static_branch_unlikely(&netstamp_needed_key) || cond) diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h index 153960663ce4..5af6eb14c5db 100644 --- a/include/net/inet_frag.h +++ b/include/net/inet_frag.h @@ -76,7 +76,7 @@ struct frag_v6_compare_key { * @stamp: timestamp of the last received fragment * @len: total length of the original datagram * @meat: length of received fragments so far - * @mono_delivery_time: stamp has a mono delivery time (EDT) + * @tstamp_type: stamp has a mono delivery time (EDT) * @flags: fragment queue flags * @max_size: maximum received fragment size * @fqdir: pointer to struct fqdir @@ -97,7 +97,7 @@ struct inet_frag_queue { ktime_t stamp; int len; int meat; - u8 mono_delivery_time; + u8 tstamp_type; __u8 flags; u16 max_size; struct fqdir *fqdir; diff --git a/net/bridge/netfilter/nf_conntrack_bridge.c b/net/bridge/netfil= ter/nf_conntrack_bridge.c index c3c51b9a6826..816bb0fde718 100644 --- a/net/bridge/netfilter/nf_conntrack_bridge.c +++ b/net/bridge/netfilter/nf_conntrack_bridge.c @@ -32,7 +32,7 @@ static int nf_br_ip_fragment(struct net *net, struct sock= *sk, struct sk_buff *)) { int frag_max_size =3D BR_INPUT_SKB_CB(skb)->frag_max_size; - bool mono_delivery_time =3D skb->mono_delivery_time; + u8 tstamp_type =3D skb->tstamp_type; unsigned int hlen, ll_rs, mtu; ktime_t tstamp =3D skb->tstamp; struct ip_frag_state state; @@ -82,7 +82,7 @@ static int nf_br_ip_fragment(struct net *net, struct sock= *sk, if (iter.frag) ip_fraglist_prepare(skb, &iter); =20 - skb_set_delivery_time(skb, tstamp, mono_delivery_time); + skb_set_delivery_time(skb, tstamp, tstamp_type); err =3D output(net, sk, data, skb); if (err || !iter.frag) break; @@ -113,7 +113,7 @@ static int nf_br_ip_fragment(struct net *net, struct so= ck *sk, goto blackhole; } =20 - skb_set_delivery_time(skb2, tstamp, mono_delivery_time); + skb_set_delivery_time(skb2, tstamp, tstamp_type); err =3D output(net, sk, data, skb2); if (err) goto blackhole; diff --git a/net/core/dev.c b/net/core/dev.c index 854a3a28a8d8..4e7dfefbb824 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2146,7 +2146,7 @@ EXPORT_SYMBOL(net_disable_timestamp); static inline void net_timestamp_set(struct sk_buff *skb) { skb->tstamp =3D 0; - skb->mono_delivery_time =3D 0; + skb->tstamp_type =3D SKB_CLOCK_REAL; if (static_branch_unlikely(&netstamp_needed_key)) skb->tstamp =3D ktime_get_real(); } diff --git a/net/core/filter.c b/net/core/filter.c index 8d185d99a643..b1192e672cbb 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -7709,13 +7709,13 @@ BPF_CALL_3(bpf_skb_set_tstamp, struct sk_buff *, sk= b, if (!tstamp) return -EINVAL; skb->tstamp =3D tstamp; - skb->mono_delivery_time =3D 1; + skb->tstamp_type =3D SKB_CLOCK_MONO; break; case BPF_SKB_TSTAMP_UNSPEC: if (tstamp) return -EINVAL; skb->tstamp =3D 0; - skb->mono_delivery_time =3D 0; + skb->tstamp_type =3D SKB_CLOCK_REAL; break; default: return -EINVAL; @@ -9422,7 +9422,7 @@ static struct bpf_insn *bpf_convert_tstamp_read(const= struct bpf_prog *prog, TC_AT_INGRESS_MASK | SKB_MONO_DELIVERY_TIME_MASK); *insn++ =3D BPF_JMP32_IMM(BPF_JNE, tmp_reg, TC_AT_INGRESS_MASK | SKB_MONO_DELIVERY_TIME_MASK, 2); - /* skb->tc_at_ingress && skb->mono_delivery_time, + /* skb->tc_at_ingress && skb->tstamp_type:1, * read 0 as the (rcv) timestamp. */ *insn++ =3D BPF_MOV64_IMM(value_reg, 0); @@ -9447,7 +9447,7 @@ static struct bpf_insn *bpf_convert_tstamp_write(cons= t struct bpf_prog *prog, * the bpf prog is aware the tstamp could have delivery time. * Thus, write skb->tstamp as is if tstamp_type_access is true. * Otherwise, writing at ingress will have to clear the - * mono_delivery_time bit also. + * mono_delivery_time (skb->tstamp_type:1)bit also. */ if (!prog->tstamp_type_access) { __u8 tmp_reg =3D BPF_REG_AX; @@ -9457,7 +9457,7 @@ static struct bpf_insn *bpf_convert_tstamp_write(cons= t struct bpf_prog *prog, *insn++ =3D BPF_JMP32_IMM(BPF_JSET, tmp_reg, TC_AT_INGRESS_MASK, 1); /* goto */ *insn++ =3D BPF_JMP_A(2); - /* : mono_delivery_time */ + /* : mono_delivery_time or (skb->tstamp_type:1) */ *insn++ =3D BPF_ALU32_IMM(BPF_AND, tmp_reg, ~SKB_MONO_DELIVERY_TIME_MASK= ); *insn++ =3D BPF_STX_MEM(BPF_B, skb_reg, tmp_reg, SKB_BF_MONO_TC_OFFSET); } diff --git a/net/ieee802154/6lowpan/reassembly.c b/net/ieee802154/6lowpan/r= eassembly.c index 6dd960ec558c..ba0455ad7701 100644 --- a/net/ieee802154/6lowpan/reassembly.c +++ b/net/ieee802154/6lowpan/reassembly.c @@ -130,7 +130,7 @@ static int lowpan_frag_queue(struct lowpan_frag_queue *= fq, goto err; =20 fq->q.stamp =3D skb->tstamp; - fq->q.mono_delivery_time =3D skb->mono_delivery_time; + fq->q.tstamp_type =3D skb->tstamp_type; if (frag_type =3D=3D LOWPAN_DISPATCH_FRAG1) fq->q.flags |=3D INET_FRAG_FIRST_IN; =20 diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c index faaec92a46ac..d179a2c84222 100644 --- a/net/ipv4/inet_fragment.c +++ b/net/ipv4/inet_fragment.c @@ -619,7 +619,7 @@ void inet_frag_reasm_finish(struct inet_frag_queue *q, = struct sk_buff *head, skb_mark_not_on_list(head); head->prev =3D NULL; head->tstamp =3D q->stamp; - head->mono_delivery_time =3D q->mono_delivery_time; + head->tstamp_type =3D q->tstamp_type; =20 if (sk) refcount_add(sum_truesize - head_truesize, &sk->sk_wmem_alloc); diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index fb947d1613fe..787aa86800f5 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -355,7 +355,7 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff= *skb) qp->iif =3D dev->ifindex; =20 qp->q.stamp =3D skb->tstamp; - qp->q.mono_delivery_time =3D skb->mono_delivery_time; + qp->q.tstamp_type =3D skb->tstamp_type; qp->q.meat +=3D skb->len; qp->ecn |=3D ecn; add_frag_mem_limit(qp->q.fqdir, skb->truesize); diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 1fe794967211..f29349151203 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -764,7 +764,7 @@ int ip_do_fragment(struct net *net, struct sock *sk, st= ruct sk_buff *skb, { struct iphdr *iph; struct sk_buff *skb2; - bool mono_delivery_time =3D skb->mono_delivery_time; + u8 tstamp_type =3D skb->tstamp_type; struct rtable *rt =3D skb_rtable(skb); unsigned int mtu, hlen, ll_rs; struct ip_fraglist_iter iter; @@ -856,7 +856,7 @@ int ip_do_fragment(struct net *net, struct sock *sk, st= ruct sk_buff *skb, } } =20 - skb_set_delivery_time(skb, tstamp, mono_delivery_time); + skb_set_delivery_time(skb, tstamp, tstamp_type); err =3D output(net, sk, skb); =20 if (!err) @@ -912,7 +912,7 @@ int ip_do_fragment(struct net *net, struct sock *sk, st= ruct sk_buff *skb, /* * Put this fragment into the sending queue. */ - skb_set_delivery_time(skb2, tstamp, mono_delivery_time); + skb_set_delivery_time(skb2, tstamp, tstamp_type); err =3D output(net, sk, skb2); if (err) goto fail; @@ -1649,7 +1649,8 @@ void ip_send_unicast_reply(struct sock *sk, struct sk= _buff *skb, arg->csumoffset) =3D csum_fold(csum_add(nskb->csum, arg->csum)); nskb->ip_summed =3D CHECKSUM_NONE; - nskb->mono_delivery_time =3D !!transmit_time; + if (transmit_time) + nskb->tstamp_type =3D SKB_CLOCK_MONO; if (txhash) skb_set_hash(nskb, txhash, PKT_HASH_TYPE_L4); ip_push_pending_frames(sk, &fl4); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 61119d42b0fd..a062f88c47c3 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1300,7 +1300,7 @@ static int __tcp_transmit_skb(struct sock *sk, struct= sk_buff *skb, tp =3D tcp_sk(sk); prior_wstamp =3D tp->tcp_wstamp_ns; tp->tcp_wstamp_ns =3D max(tp->tcp_wstamp_ns, tp->tcp_clock_cache); - skb_set_delivery_time(skb, tp->tcp_wstamp_ns, true); + skb_set_delivery_time(skb, tp->tcp_wstamp_ns, CLOCK_MONOTONIC); if (clone_it) { oskb =3D skb; =20 @@ -1650,7 +1650,7 @@ int tcp_fragment(struct sock *sk, enum tcp_queue tcp_= queue, =20 skb_split(skb, buff, len); =20 - skb_set_delivery_time(buff, skb->tstamp, true); + skb_set_delivery_time(buff, skb->tstamp, CLOCK_MONOTONIC); tcp_fragment_tstamp(skb, buff); =20 old_factor =3D tcp_skb_pcount(skb); @@ -2731,7 +2731,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned = int mss_now, int nonagle, if (unlikely(tp->repair) && tp->repair_queue =3D=3D TCP_SEND_QUEUE) { /* "skb_mstamp_ns" is used as a start point for the retransmit timer */ tp->tcp_wstamp_ns =3D tp->tcp_clock_cache; - skb_set_delivery_time(skb, tp->tcp_wstamp_ns, true); + skb_set_delivery_time(skb, tp->tcp_wstamp_ns, CLOCK_MONOTONIC); list_move_tail(&skb->tcp_tsorted_anchor, &tp->tsorted_sent_queue); tcp_init_tso_segs(skb, mss_now); goto repair; /* Skip network transmission */ @@ -3714,11 +3714,11 @@ struct sk_buff *tcp_make_synack(const struct sock *= sk, struct dst_entry *dst, #ifdef CONFIG_SYN_COOKIES if (unlikely(synack_type =3D=3D TCP_SYNACK_COOKIE && ireq->tstamp_ok)) skb_set_delivery_time(skb, cookie_init_timestamp(req, now), - true); + CLOCK_MONOTONIC); else #endif { - skb_set_delivery_time(skb, now, true); + skb_set_delivery_time(skb, now, CLOCK_MONOTONIC); if (!tcp_rsk(req)->snt_synack) /* Timestamp first SYNACK */ tcp_rsk(req)->snt_synack =3D tcp_skb_timestamp_us(skb); } @@ -3805,7 +3805,7 @@ struct sk_buff *tcp_make_synack(const struct sock *sk= , struct dst_entry *dst, bpf_skops_write_hdr_opt((struct sock *)sk, skb, req, syn_skb, synack_type, &opts); =20 - skb_set_delivery_time(skb, now, true); + skb_set_delivery_time(skb, now, CLOCK_MONOTONIC); tcp_add_tx_delay(skb, tp); =20 return skb; @@ -3989,7 +3989,7 @@ static int tcp_send_syn_data(struct sock *sk, struct = sk_buff *syn) =20 err =3D tcp_transmit_skb(sk, syn_data, 1, sk->sk_allocation); =20 - skb_set_delivery_time(syn, syn_data->skb_mstamp_ns, true); + skb_set_delivery_time(syn, syn_data->skb_mstamp_ns, CLOCK_MONOTONIC); =20 /* Now full SYN+DATA was cloned and sent (or not), * remove the SYN from the original skb (syn_data) diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index b9dd3a66e423..a9e819115622 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -859,7 +859,7 @@ int ip6_fragment(struct net *net, struct sock *sk, stru= ct sk_buff *skb, struct rt6_info *rt =3D (struct rt6_info *)skb_dst(skb); struct ipv6_pinfo *np =3D skb->sk && !dev_recursion_level() ? inet6_sk(skb->sk) : NULL; - bool mono_delivery_time =3D skb->mono_delivery_time; + u8 tstamp_type =3D skb->tstamp_type; struct ip6_frag_state state; unsigned int mtu, hlen, nexthdr_offset; ktime_t tstamp =3D skb->tstamp; @@ -955,7 +955,7 @@ int ip6_fragment(struct net *net, struct sock *sk, stru= ct sk_buff *skb, if (iter.frag) ip6_fraglist_prepare(skb, &iter); =20 - skb_set_delivery_time(skb, tstamp, mono_delivery_time); + skb_set_delivery_time(skb, tstamp, tstamp_type); err =3D output(net, sk, skb); if (!err) IP6_INC_STATS(net, ip6_dst_idev(&rt->dst), @@ -1016,7 +1016,7 @@ int ip6_fragment(struct net *net, struct sock *sk, st= ruct sk_buff *skb, /* * Put this fragment into the sending queue. */ - skb_set_delivery_time(frag, tstamp, mono_delivery_time); + skb_set_delivery_time(frag, tstamp, tstamp_type); err =3D output(net, sk, frag); if (err) goto fail; diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c index 53d255838e6a..e0c2347b4dc6 100644 --- a/net/ipv6/netfilter.c +++ b/net/ipv6/netfilter.c @@ -126,7 +126,7 @@ int br_ip6_fragment(struct net *net, struct sock *sk, s= truct sk_buff *skb, struct sk_buff *)) { int frag_max_size =3D BR_INPUT_SKB_CB(skb)->frag_max_size; - bool mono_delivery_time =3D skb->mono_delivery_time; + u8 tstamp_type =3D skb->tstamp_type; ktime_t tstamp =3D skb->tstamp; struct ip6_frag_state state; u8 *prevhdr, nexthdr =3D 0; @@ -192,7 +192,7 @@ int br_ip6_fragment(struct net *net, struct sock *sk, s= truct sk_buff *skb, if (iter.frag) ip6_fraglist_prepare(skb, &iter); =20 - skb_set_delivery_time(skb, tstamp, mono_delivery_time); + skb_set_delivery_time(skb, tstamp, tstamp_type); err =3D output(net, sk, data, skb); if (err || !iter.frag) break; @@ -225,7 +225,7 @@ int br_ip6_fragment(struct net *net, struct sock *sk, s= truct sk_buff *skb, goto blackhole; } =20 - skb_set_delivery_time(skb2, tstamp, mono_delivery_time); + skb_set_delivery_time(skb2, tstamp, tstamp_type); err =3D output(net, sk, data, skb2); if (err) goto blackhole; diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/n= f_conntrack_reasm.c index d0dcbaca1994..5cc5d823d33f 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -264,7 +264,7 @@ static int nf_ct_frag6_queue(struct frag_queue *fq, str= uct sk_buff *skb, fq->iif =3D dev->ifindex; =20 fq->q.stamp =3D skb->tstamp; - fq->q.mono_delivery_time =3D skb->mono_delivery_time; + fq->q.tstamp_type =3D skb->tstamp_type; fq->q.meat +=3D skb->len; fq->ecn |=3D ecn; if (payload_len > fq->q.max_size) diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index acb4f119e11f..ea724ff558b4 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -198,7 +198,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct= sk_buff *skb, fq->iif =3D dev->ifindex; =20 fq->q.stamp =3D skb->tstamp; - fq->q.mono_delivery_time =3D skb->mono_delivery_time; + fq->q.tstamp_type =3D skb->tstamp_type; fq->q.meat +=3D skb->len; fq->ecn |=3D ecn; add_frag_mem_limit(fq->q.fqdir, skb->truesize); diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index bb7c3caf4f85..7f8a8bd77547 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -975,7 +975,7 @@ static void tcp_v6_send_response(const struct sock *sk,= struct sk_buff *skb, u32 mark =3D inet_twsk(sk)->tw_mark; else mark =3D READ_ONCE(sk->sk_mark); - skb_set_delivery_time(buff, tcp_transmit_time(sk), true); + skb_set_delivery_time(buff, tcp_transmit_time(sk), CLOCK_MONOTONIC); } if (txhash) { /* autoflowlabel/skb_get_hash_flowi6 rely on buff->hash */ diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c index 0e3cf11ae5fc..9cd12d61818a 100644 --- a/net/sched/act_bpf.c +++ b/net/sched/act_bpf.c @@ -54,8 +54,8 @@ TC_INDIRECT_SCOPE int tcf_bpf_act(struct sk_buff *skb, bpf_compute_data_pointers(skb); filter_res =3D bpf_prog_run(filter, skb); } - if (unlikely(!skb->tstamp && skb->mono_delivery_time)) - skb->mono_delivery_time =3D 0; + if (unlikely(!skb->tstamp && skb->tstamp_type)) + skb->tstamp_type =3D SKB_CLOCK_REAL; if (skb_sk_is_prefetched(skb) && filter_res !=3D TC_ACT_OK) skb_orphan(skb); =20 diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c index 5e83e890f6a4..506516be5287 100644 --- a/net/sched/cls_bpf.c +++ b/net/sched/cls_bpf.c @@ -104,8 +104,8 @@ TC_INDIRECT_SCOPE int cls_bpf_classify(struct sk_buff *= skb, bpf_compute_data_pointers(skb); filter_res =3D bpf_prog_run(prog->filter, skb); } - if (unlikely(!skb->tstamp && skb->mono_delivery_time)) - skb->mono_delivery_time =3D 0; + if (unlikely(!skb->tstamp && skb->tstamp_type)) + skb->tstamp_type =3D SKB_CLOCK_REAL; =20 if (prog->exts_integrated) { res->class =3D 0; --=20 2.25.1 From nobody Wed May 22 02:33:18 2024 Received: from mx0b-0031df01.pphosted.com (mx0b-0031df01.pphosted.com [205.220.180.131]) (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 A5E1E817; Thu, 18 Apr 2024 00:43:32 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=205.220.180.131 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1713401014; cv=none; b=HkBbp2oEbmOFI19H9A7B0ozudfOf6ezFRw5hHE45HDZyNJzmNTcj3vrQkKmUrfCTBiRJD01P5njDzfyffBEZZAfPcqk2I06AgqJTXufzEOhVKBmL17U0ZTGIHdjdUop05PCxgVyER3vr4DKm3LSBFoitF+AwQBsYzBHWKmngWRY= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1713401014; c=relaxed/simple; bh=K9IrT5hk5i1L5atc0AX3jnLghQtoms5HfxM8mxo6hSg=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=Fbr9cIvTusVDURWt2KgiFh2lNOtAQL3EF/zGqVB3wikLtUhelFNYEwPQ6p2/yGvskd9pYYuexXDQdxEnv5KFXfL+Hqhz8hmzbv6wi2j0kAg4rKysgeWeDaY1ac0ApFxD01uP+pj3EeGZojZnkHaU1AYVxqab/9byD2CPg8VHxAY= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=quicinc.com; spf=pass smtp.mailfrom=qualcomm.com; dkim=pass (2048-bit key) header.d=quicinc.com header.i=@quicinc.com header.b=O3x8Zr+q; arc=none smtp.client-ip=205.220.180.131 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=quicinc.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=qualcomm.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=quicinc.com header.i=@quicinc.com header.b="O3x8Zr+q" Received: from pps.filterd (m0279870.ppops.net [127.0.0.1]) by mx0a-0031df01.pphosted.com (8.17.1.24/8.17.1.24) with ESMTP id 43HNmbek015743; Thu, 18 Apr 2024 00:43:11 GMT DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=quicinc.com; h= from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; s=qcppdkim1; bh=H9rH3ft F5DX0Qysq72cAS8A5jzstsSbXbMANg6pF9sQ=; b=O3x8Zr+qybt7cq1Mvc69Dic cDEZkBxbtPu5NtxvTJTHNpzZuG+M3nJ9RH/Eoj/+vyn6jzrGSMYaf3/TXnYTBD44 fqxBemlSM6pD25nV5NvnAs0Ibi50ZujrgkNqG+p4P9yWyCW/rpsZBTjupiQQ5wvu WVi17JU337DZMY4uZa8g5kXGT4kPVw7XqEvc8N+RezScd1Ea//9qpaenusqxRBQc cHJn3Rsk18XYC0CnTlFxCU3b/N+McGyUYaGWDhlUfLi7SLw9mevHZnWrVY0z4GUv A/4ozwCpwOzXx0R51Pcd427ImZxg8rvUOTpAQ1S76T7BpU7EtZLafxqNEPE1vZA= = Received: from nalasppmta04.qualcomm.com (Global_NAT1.qualcomm.com [129.46.96.20]) by mx0a-0031df01.pphosted.com (PPS) with ESMTPS id 3xjrdbg3a8-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Thu, 18 Apr 2024 00:43:11 +0000 (GMT) Received: from pps.filterd (NALASPPMTA04.qualcomm.com [127.0.0.1]) by NALASPPMTA04.qualcomm.com (8.17.1.5/8.17.1.5) with ESMTP id 43I0g5Hb023763; Thu, 18 Apr 2024 00:43:09 GMT Received: from pps.reinject (localhost [127.0.0.1]) by NALASPPMTA04.qualcomm.com (PPS) with ESMTPS id 3xj8myfdpj-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Thu, 18 Apr 2024 00:43:09 +0000 Received: from NALASPPMTA04.qualcomm.com (NALASPPMTA04.qualcomm.com [127.0.0.1]) by pps.reinject (8.17.1.5/8.17.1.5) with ESMTP id 43I0f9Zk022970; Thu, 18 Apr 2024 00:43:09 GMT Received: from hu-devc-lv-u20-a-new.qualcomm.com (hu-abchauha-lv.qualcomm.com [10.81.25.35]) by NALASPPMTA04.qualcomm.com (PPS) with ESMTPS id 43I0h9II026807 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Thu, 18 Apr 2024 00:43:09 +0000 Received: by hu-devc-lv-u20-a-new.qualcomm.com (Postfix, from userid 214165) id BAB5E23975; Wed, 17 Apr 2024 17:43:08 -0700 (PDT) From: Abhishek Chauhan To: "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , netdev@vger.kernel.org, linux-kernel@vger.kernel.org, Andrew Halaney , Willem de Bruijn , Martin KaFai Lau , Martin KaFai Lau , Daniel Borkmann , bpf Cc: kernel@quicinc.com Subject: [RFC PATCH bpf-next v4 2/2] net: Add additional bit to support clockid_t timestamp type Date: Wed, 17 Apr 2024 17:43:08 -0700 Message-Id: <20240418004308.1009262-3-quic_abchauha@quicinc.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20240418004308.1009262-1-quic_abchauha@quicinc.com> References: <20240418004308.1009262-1-quic_abchauha@quicinc.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-QCInternal: smtphost X-QCInternal: smtphost X-Proofpoint-Virus-Version: vendor=nai engine=6200 definitions=5800 signatures=585085 X-Proofpoint-Virus-Version: vendor=nai engine=6200 definitions=5800 signatures=585085 X-Proofpoint-GUID: 7SWxrjnE-QCXaILwEvxVq2JernteLvvB X-Proofpoint-ORIG-GUID: 7SWxrjnE-QCXaILwEvxVq2JernteLvvB X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.272,Aquarius:18.0.1011,Hydra:6.0.619,FMLib:17.11.176.26 definitions=2024-04-17_19,2024-04-17_01,2023-05-22_02 X-Proofpoint-Spam-Details: rule=outbound_notspam policy=outbound score=0 phishscore=0 impostorscore=0 bulkscore=0 lowpriorityscore=0 clxscore=1015 adultscore=0 malwarescore=0 spamscore=0 priorityscore=1501 suspectscore=0 mlxlogscore=999 mlxscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.19.0-2404010003 definitions=main-2404180001 Content-Type: text/plain; charset="utf-8" tstamp_type is now set based on actual clockid_t compressed into 2 bits. To make the design scalable for future needs this commit bring in the change to extend the tstamp_type:1 to tstamp_type:2 to support other clockid_t timestamp. We now support CLOCK_TAI as part of tstamp_type as part of this commit with exisiting support CLOCK_MONOTONIC and CLOCK_REALTIME. Link: https://lore.kernel.org/netdev/bc037db4-58bb-4861-ac31-a361a93841d3@l= inux.dev/ Signed-off-by: Abhishek Chauhan --- Changes since v3 - Carefully reviewed BPF APIs and made changes in=20 BPF code as well.=20 - Re-used actual clockid_t values since skbuff.h=20 indirectly includes uapi/linux/time.h - Added CLOCK_TAI as part of the skb_set_delivery_time handling instead of CLOCK_USER - Added default in switch for unsupported and invalid=20 timestamp with an WARN_ONCE. - All of the above comments were given by Willem =20 - Made changes in filter.c as per Martin's comments to handle invalid cases in bpf code with addition of SKB_TAI_DELIVERY_TIME_MASK Changes since v2 - Minor changes to commit subject Changes since v1=20 - identified additional changes in BPF framework. - Bit shift in SKB_MONO_DELIVERY_TIME_MASK and TC_AT_INGRESS_MASK. - Made changes in skb_set_delivery_time to keep changes similar to=20 previous code for mono_delivery_time and just setting tstamp_type bit 1 for userspace timestamp. include/linux/skbuff.h | 21 +++++++--- include/uapi/linux/bpf.h | 1 + net/core/filter.c | 40 ++++++++++++++++--- net/ipv4/ip_output.c | 2 +- net/ipv4/raw.c | 2 +- net/ipv6/ip6_output.c | 2 +- net/ipv6/raw.c | 2 +- net/packet/af_packet.c | 7 ++-- .../selftests/bpf/prog_tests/ctx_rewrite.c | 13 ++++-- 9 files changed, 68 insertions(+), 22 deletions(-) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 2e7811c7e4db..647522ef41cd 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -703,14 +703,17 @@ typedef unsigned char *sk_buff_data_t; #endif =20 /** - * tstamp_type:1 can take 2 values each + * tstamp_type:2 can take 4 values each * represented by time base in skb * 0x0 =3D> real timestamp_type * 0x1 =3D> mono timestamp_type + * 0x2 =3D> tai timestamp_type + * 0x3 =3D> undefined timestamp_type */ enum skb_tstamp_type { SKB_CLOCK_REAL, /* Time base is skb is REALTIME */ SKB_CLOCK_MONO, /* Time base is skb is MONOTONIC */ + SKB_CLOCK_TAI, /* Time base in skb is TAI */ }; =20 /** @@ -833,7 +836,8 @@ enum skb_tstamp_type { * @tstamp_type: When set, skb->tstamp has the * delivery_time in mono clock base (i.e. EDT). Otherwise, the * skb->tstamp has the (rcv) timestamp at ingress and - * delivery_time at egress. + * delivery_time at egress or skb->tstamp defined by skb->sk->sk_clockid + * coming from userspace * @napi_id: id of the NAPI struct this skb came from * @sender_cpu: (aka @napi_id) source CPU in XPS * @alloc_cpu: CPU which did the skb allocation. @@ -961,7 +965,7 @@ struct sk_buff { /* private: */ __u8 __mono_tc_offset[0]; /* public: */ - __u8 tstamp_type:1; /* See SKB_CLOCK_*_MASK */ + __u8 tstamp_type:2; /* See skb_tstamp_type enum */ #ifdef CONFIG_NET_XGRESS __u8 tc_at_ingress:1; /* See TC_AT_INGRESS_MASK */ __u8 tc_skip_classify:1; @@ -1096,10 +1100,12 @@ struct sk_buff { */ #ifdef __BIG_ENDIAN_BITFIELD #define SKB_MONO_DELIVERY_TIME_MASK (1 << 7) -#define TC_AT_INGRESS_MASK (1 << 6) +#define SKB_TAI_DELIVERY_TIME_MASK (1 << 6) +#define TC_AT_INGRESS_MASK (1 << 5) #else #define SKB_MONO_DELIVERY_TIME_MASK (1 << 0) -#define TC_AT_INGRESS_MASK (1 << 1) +#define SKB_TAI_DELIVERY_TIME_MASK (1 << 1) +#define TC_AT_INGRESS_MASK (1 << 2) #endif #define SKB_BF_MONO_TC_OFFSET offsetof(struct sk_buff, __mono_tc_offset) =20 @@ -4206,6 +4212,11 @@ static inline void skb_set_delivery_time(struct sk_b= uff *skb, ktime_t kt, case CLOCK_MONOTONIC: skb->tstamp_type =3D SKB_CLOCK_MONO; break; + case CLOCK_TAI: + skb->tstamp_type =3D SKB_CLOCK_TAI; + break; + default: + WARN_ONCE(true, "clockid %d not supported", tstamp_type); } } =20 diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index cee0a7915c08..1376ed5ece10 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -6209,6 +6209,7 @@ union { \ enum { BPF_SKB_TSTAMP_UNSPEC, BPF_SKB_TSTAMP_DELIVERY_MONO, /* tstamp has mono delivery time */ + BPF_SKB_TSTAMP_DELIVERY_TAI, /* tstamp has tai delivery time */ /* For any BPF_SKB_TSTAMP_* that the bpf prog cannot handle, * the bpf prog should handle it like BPF_SKB_TSTAMP_UNSPEC * and try to deduce it by ingress, egress or skb->sk->sk_clockid. diff --git a/net/core/filter.c b/net/core/filter.c index b1192e672cbb..de86c42b5859 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -7711,6 +7711,12 @@ BPF_CALL_3(bpf_skb_set_tstamp, struct sk_buff *, skb, skb->tstamp =3D tstamp; skb->tstamp_type =3D SKB_CLOCK_MONO; break; + case BPF_SKB_TSTAMP_DELIVERY_TAI: + if (!tstamp) + return -EINVAL; + skb->tstamp =3D tstamp; + skb->tstamp_type =3D SKB_CLOCK_TAI; + break; case BPF_SKB_TSTAMP_UNSPEC: if (tstamp) return -EINVAL; @@ -9372,10 +9378,16 @@ static struct bpf_insn *bpf_convert_tstamp_type_rea= d(const struct bpf_insn *si, *insn++ =3D BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg, SKB_BF_MONO_TC_OFFSET); *insn++ =3D BPF_JMP32_IMM(BPF_JSET, tmp_reg, - SKB_MONO_DELIVERY_TIME_MASK, 2); + SKB_MONO_DELIVERY_TIME_MASK | SKB_TAI_DELIVERY_TIME_MASK, 2); + *insn++ =3D BPF_JMP32_IMM(BPF_JSET, tmp_reg, + SKB_MONO_DELIVERY_TIME_MASK, 3); + *insn++ =3D BPF_JMP32_IMM(BPF_JSET, tmp_reg, + SKB_TAI_DELIVERY_TIME_MASK, 4); *insn++ =3D BPF_MOV32_IMM(value_reg, BPF_SKB_TSTAMP_UNSPEC); *insn++ =3D BPF_JMP_A(1); *insn++ =3D BPF_MOV32_IMM(value_reg, BPF_SKB_TSTAMP_DELIVERY_MONO); + *insn++ =3D BPF_JMP_A(1); + *insn++ =3D BPF_MOV32_IMM(value_reg, BPF_SKB_TSTAMP_DELIVERY_TAI); =20 return insn; } @@ -9418,10 +9430,26 @@ static struct bpf_insn *bpf_convert_tstamp_read(con= st struct bpf_prog *prog, __u8 tmp_reg =3D BPF_REG_AX; =20 *insn++ =3D BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg, SKB_BF_MONO_TC_OFFSET); + /*check if all three bits are set*/ *insn++ =3D BPF_ALU32_IMM(BPF_AND, tmp_reg, - TC_AT_INGRESS_MASK | SKB_MONO_DELIVERY_TIME_MASK); - *insn++ =3D BPF_JMP32_IMM(BPF_JNE, tmp_reg, - TC_AT_INGRESS_MASK | SKB_MONO_DELIVERY_TIME_MASK, 2); + TC_AT_INGRESS_MASK | SKB_MONO_DELIVERY_TIME_MASK | + SKB_TAI_DELIVERY_TIME_MASK); + /*if all 3 bits are set jump 3 instructions and clear the register */ + *insn++ =3D BPF_JMP32_IMM(BPF_JEQ, tmp_reg, + TC_AT_INGRESS_MASK | SKB_MONO_DELIVERY_TIME_MASK | + SKB_TAI_DELIVERY_TIME_MASK, 4); + /*Now check Mono is set with ingress mask if so clear */ + *insn++ =3D BPF_JMP32_IMM(BPF_JEQ, tmp_reg, + TC_AT_INGRESS_MASK | SKB_MONO_DELIVERY_TIME_MASK, 3); + /*Now Check tai is set with ingress mask if so clear */ + *insn++ =3D BPF_JMP32_IMM(BPF_JEQ, tmp_reg, + TC_AT_INGRESS_MASK | SKB_TAI_DELIVERY_TIME_MASK, 2); + /*Now Check tai and mono are set if so clear */ + *insn++ =3D BPF_JMP32_IMM(BPF_JEQ, tmp_reg, + SKB_MONO_DELIVERY_TIME_MASK | + SKB_TAI_DELIVERY_TIME_MASK, 1); + /* goto */ + *insn++ =3D BPF_JMP_A(2); /* skb->tc_at_ingress && skb->tstamp_type:1, * read 0 as the (rcv) timestamp. */ @@ -9456,9 +9484,11 @@ static struct bpf_insn *bpf_convert_tstamp_write(con= st struct bpf_prog *prog, /* Writing __sk_buff->tstamp as ingress, goto */ *insn++ =3D BPF_JMP32_IMM(BPF_JSET, tmp_reg, TC_AT_INGRESS_MASK, 1); /* goto */ - *insn++ =3D BPF_JMP_A(2); + *insn++ =3D BPF_JMP_A(3); /* : mono_delivery_time or (skb->tstamp_type:1) */ *insn++ =3D BPF_ALU32_IMM(BPF_AND, tmp_reg, ~SKB_MONO_DELIVERY_TIME_MASK= ); + /* : tai delivery_time or (skb->tstamp_type:2) */ + *insn++ =3D BPF_ALU32_IMM(BPF_AND, tmp_reg, ~SKB_TAI_DELIVERY_TIME_MASK); *insn++ =3D BPF_STX_MEM(BPF_B, skb_reg, tmp_reg, SKB_BF_MONO_TC_OFFSET); } #endif diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index f29349151203..91e80a3d411f 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -1457,7 +1457,7 @@ struct sk_buff *__ip_make_skb(struct sock *sk, =20 skb->priority =3D (cork->tos !=3D -1) ? cork->priority: READ_ONCE(sk->sk_= priority); skb->mark =3D cork->mark; - skb->tstamp =3D cork->transmit_time; + skb_set_delivery_time(skb, cork->transmit_time, sk->sk_clockid); /* * Steal rt from cork.dst to avoid a pair of atomic_inc/atomic_dec * on dst refcount diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index dcb11f22cbf2..bbc46a40c8b6 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -360,7 +360,7 @@ static int raw_send_hdrinc(struct sock *sk, struct flow= i4 *fl4, skb->protocol =3D htons(ETH_P_IP); skb->priority =3D READ_ONCE(sk->sk_priority); skb->mark =3D sockc->mark; - skb->tstamp =3D sockc->transmit_time; + skb_set_delivery_time(skb, sockc->transmit_time, sk->sk_clockid); skb_dst_set(skb, &rt->dst); *rtp =3D NULL; =20 diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index a9e819115622..0b8193bdd98f 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -1924,7 +1924,7 @@ struct sk_buff *__ip6_make_skb(struct sock *sk, =20 skb->priority =3D READ_ONCE(sk->sk_priority); skb->mark =3D cork->base.mark; - skb->tstamp =3D cork->base.transmit_time; + skb_set_delivery_time(skb, cork->base.transmit_time, sk->sk_clockid); =20 ip6_cork_steal_dst(skb, cork); IP6_INC_STATS(net, rt->rt6i_idev, IPSTATS_MIB_OUTREQUESTS); diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 0d896ca7b589..625f3a917e50 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -621,7 +621,7 @@ static int rawv6_send_hdrinc(struct sock *sk, struct ms= ghdr *msg, int length, skb->protocol =3D htons(ETH_P_IPV6); skb->priority =3D READ_ONCE(sk->sk_priority); skb->mark =3D sockc->mark; - skb->tstamp =3D sockc->transmit_time; + skb_set_delivery_time(skb, sockc->transmit_time, sk->sk_clockid); =20 skb_put(skb, length); skb_reset_network_header(skb); diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 8c6d3fbb4ed8..356c96f23370 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -2056,8 +2056,7 @@ static int packet_sendmsg_spkt(struct socket *sock, s= truct msghdr *msg, skb->dev =3D dev; skb->priority =3D READ_ONCE(sk->sk_priority); skb->mark =3D READ_ONCE(sk->sk_mark); - skb->tstamp =3D sockc.transmit_time; - + skb_set_delivery_time(skb, sockc.transmit_time, sk->sk_clockid); skb_setup_tx_timestamp(skb, sockc.tsflags); =20 if (unlikely(extra_len =3D=3D 4)) @@ -2585,7 +2584,7 @@ static int tpacket_fill_skb(struct packet_sock *po, s= truct sk_buff *skb, skb->dev =3D dev; skb->priority =3D READ_ONCE(po->sk.sk_priority); skb->mark =3D READ_ONCE(po->sk.sk_mark); - skb->tstamp =3D sockc->transmit_time; + skb_set_delivery_time(skb, sockc->transmit_time, po->sk.sk_clockid); skb_setup_tx_timestamp(skb, sockc->tsflags); skb_zcopy_set_nouarg(skb, ph.raw); =20 @@ -3063,7 +3062,7 @@ static int packet_snd(struct socket *sock, struct msg= hdr *msg, size_t len) skb->dev =3D dev; skb->priority =3D READ_ONCE(sk->sk_priority); skb->mark =3D sockc.mark; - skb->tstamp =3D sockc.transmit_time; + skb_set_delivery_time(skb, sockc.transmit_time, sk->sk_clockid); =20 if (unlikely(extra_len =3D=3D 4)) skb->no_fcs =3D 1; diff --git a/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c b/tools/t= esting/selftests/bpf/prog_tests/ctx_rewrite.c index 3b7c57fe55a5..9d0c33d7f52e 100644 --- a/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c +++ b/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c @@ -69,15 +69,20 @@ static struct test_case test_cases[] =3D { { N(SCHED_CLS, struct __sk_buff, tstamp), .read =3D "r11 =3D *(u8 *)($ctx + sk_buff::__mono_tc_offset);" - "w11 &=3D 3;" - "if w11 !=3D 0x3 goto pc+2;" + "w11 &=3D 7;" + "if w11 =3D=3D 0x7 goto pc+4;" + "if w11 =3D=3D 0x5 goto pc+3;" + "if w11 =3D=3D 0x6 goto pc+2;" + "if w11 =3D=3D 0x3 goto pc+1;" + "goto pc+2" "$dst =3D 0;" "goto pc+1;" "$dst =3D *(u64 *)($ctx + sk_buff::tstamp);", .write =3D "r11 =3D *(u8 *)($ctx + sk_buff::__mono_tc_offset);" - "if w11 & 0x2 goto pc+1;" - "goto pc+2;" + "if w11 & 0x4 goto pc+1;" + "goto pc+3;" "w11 &=3D -2;" + "w11 &=3D -3;" "*(u8 *)($ctx + sk_buff::__mono_tc_offset) =3D r11;" "*(u64 *)($ctx + sk_buff::tstamp) =3D $src;", }, --=20 2.25.1