BPF List
 help / color / mirror / Atom feed
From: Martin KaFai Lau <kafai@fb.com>
To: <bpf@vger.kernel.org>, <netdev@vger.kernel.org>
Cc: Alexei Starovoitov <ast@kernel.org>,
	Andrii Nakryiko <andrii@kernel.org>,
	Daniel Borkmann <daniel@iogearbox.net>,
	David Miller <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	Jakub Kicinski <kuba@kernel.org>, <kernel-team@fb.com>,
	Willem de Bruijn <willemb@google.com>
Subject: [PATCH v6 net-next 11/13] bpf: Keep the (rcv) timestamp behavior for the existing tc-bpf@ingress
Date: Wed, 2 Mar 2022 11:56:28 -0800	[thread overview]
Message-ID: <20220302195628.3484598-1-kafai@fb.com> (raw)
In-Reply-To: <20220302195519.3479274-1-kafai@fb.com>

The current tc-bpf@ingress reads and writes the __sk_buff->tstamp
as a (rcv) timestamp which currently could either be 0 (not available)
or ktime_get_real().  This patch is to backward compatible with the
(rcv) timestamp expectation at ingress.  If the skb->tstamp has
the delivery_time, the bpf insn rewrite will read 0 for tc-bpf
running at ingress as it is not available.  When writing at ingress,
it will also clear the skb->mono_delivery_time bit.

/* BPF_READ: a = __sk_buff->tstamp */
if (!skb->tc_at_ingress || !skb->mono_delivery_time)
	a = skb->tstamp;
else
	a = 0

/* BPF_WRITE: __sk_buff->tstamp = a */
if (skb->tc_at_ingress)
	skb->mono_delivery_time = 0;
skb->tstamp = a;

[ A note on the BPF_CGROUP_INET_INGRESS which can also access
  skb->tstamp.  At that point, the skb is delivered locally
  and skb_clear_delivery_time() has already been done,
  so the skb->tstamp will only have the (rcv) timestamp. ]

If the tc-bpf@egress writes 0 to skb->tstamp, the skb->mono_delivery_time
has to be cleared also.  It could be done together during
convert_ctx_access().  However, the latter patch will also expose
the skb->mono_delivery_time bit as __sk_buff->delivery_time_type.
Changing the delivery_time_type in the background may surprise
the user, e.g. the 2nd read on __sk_buff->delivery_time_type
may need a READ_ONCE() to avoid compiler optimization.  Thus,
in expecting the needs in the latter patch, this patch does a
check on !skb->tstamp after running the tc-bpf and clears the
skb->mono_delivery_time bit if needed.  The earlier discussion
on v4 [0].

The bpf insn rewrite requires the skb's mono_delivery_time bit and
tc_at_ingress bit.  They are moved up in sk_buff so that bpf rewrite
can be done at a fixed offset.  tc_skip_classify is moved together with
tc_at_ingress.  To get one bit for mono_delivery_time, csum_not_inet is
moved down and this bit is currently used by sctp.

[0]: https://lore.kernel.org/bpf/20220217015043.khqwqklx45c4m4se@kafai-mbp.dhcp.thefacebook.com/

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
---
 include/linux/skbuff.h | 18 +++++++----
 net/core/filter.c      | 71 ++++++++++++++++++++++++++++++++++++------
 net/sched/act_bpf.c    |  2 ++
 net/sched/cls_bpf.c    |  2 ++
 4 files changed, 77 insertions(+), 16 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 4b5b926a81f2..5445860e1ba6 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -941,8 +941,12 @@ struct sk_buff {
 	__u8			vlan_present:1;	/* See PKT_VLAN_PRESENT_BIT */
 	__u8			csum_complete_sw:1;
 	__u8			csum_level:2;
-	__u8			csum_not_inet:1;
 	__u8			dst_pending_confirm:1;
+	__u8			mono_delivery_time:1;
+#ifdef CONFIG_NET_CLS_ACT
+	__u8			tc_skip_classify:1;
+	__u8			tc_at_ingress:1;
+#endif
 #ifdef CONFIG_IPV6_NDISC_NODETYPE
 	__u8			ndisc_nodetype:2;
 #endif
@@ -953,10 +957,6 @@ struct sk_buff {
 #ifdef CONFIG_NET_SWITCHDEV
 	__u8			offload_fwd_mark:1;
 	__u8			offload_l3_fwd_mark:1;
-#endif
-#ifdef CONFIG_NET_CLS_ACT
-	__u8			tc_skip_classify:1;
-	__u8			tc_at_ingress:1;
 #endif
 	__u8			redirected:1;
 #ifdef CONFIG_NET_REDIRECT
@@ -969,7 +969,7 @@ struct sk_buff {
 	__u8			decrypted:1;
 #endif
 	__u8			slow_gro:1;
-	__u8			mono_delivery_time:1;
+	__u8			csum_not_inet:1;
 
 #ifdef CONFIG_NET_SCHED
 	__u16			tc_index;	/* traffic control index */
@@ -1047,10 +1047,16 @@ struct sk_buff {
 /* if you move pkt_vlan_present around you also must adapt these constants */
 #ifdef __BIG_ENDIAN_BITFIELD
 #define PKT_VLAN_PRESENT_BIT	7
+#define TC_AT_INGRESS_MASK		(1 << 0)
+#define SKB_MONO_DELIVERY_TIME_MASK	(1 << 2)
 #else
 #define PKT_VLAN_PRESENT_BIT	0
+#define TC_AT_INGRESS_MASK		(1 << 7)
+#define SKB_MONO_DELIVERY_TIME_MASK	(1 << 5)
 #endif
 #define PKT_VLAN_PRESENT_OFFSET	offsetof(struct sk_buff, __pkt_vlan_present_offset)
+#define TC_AT_INGRESS_OFFSET offsetof(struct sk_buff, __pkt_vlan_present_offset)
+#define SKB_MONO_DELIVERY_TIME_OFFSET offsetof(struct sk_buff, __pkt_vlan_present_offset)
 
 #ifdef __KERNEL__
 /*
diff --git a/net/core/filter.c b/net/core/filter.c
index cfcf9b4d1ec2..5072733743e9 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -8859,6 +8859,65 @@ static struct bpf_insn *bpf_convert_shinfo_access(const struct bpf_insn *si,
 	return insn;
 }
 
+static struct bpf_insn *bpf_convert_tstamp_read(const struct bpf_insn *si,
+						struct bpf_insn *insn)
+{
+	__u8 value_reg = si->dst_reg;
+	__u8 skb_reg = si->src_reg;
+
+#ifdef CONFIG_NET_CLS_ACT
+	__u8 tmp_reg = BPF_REG_AX;
+
+	*insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg, TC_AT_INGRESS_OFFSET);
+	*insn++ = BPF_ALU32_IMM(BPF_AND, tmp_reg, TC_AT_INGRESS_MASK);
+	*insn++ = BPF_JMP32_IMM(BPF_JEQ, tmp_reg, 0, 5);
+	/* @ingress, read __sk_buff->tstamp as the (rcv) timestamp,
+	 * so check the skb->mono_delivery_time.
+	 */
+	*insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg,
+			      SKB_MONO_DELIVERY_TIME_OFFSET);
+	*insn++ = BPF_ALU32_IMM(BPF_AND, tmp_reg,
+				SKB_MONO_DELIVERY_TIME_MASK);
+	*insn++ = BPF_JMP32_IMM(BPF_JEQ, tmp_reg, 0, 2);
+	/* skb->mono_delivery_time is set, read 0 as the (rcv) timestamp. */
+	*insn++ = BPF_MOV64_IMM(value_reg, 0);
+	*insn++ = BPF_JMP_A(1);
+#endif
+
+	*insn++ = BPF_LDX_MEM(BPF_DW, value_reg, skb_reg,
+			      offsetof(struct sk_buff, tstamp));
+	return insn;
+}
+
+static struct bpf_insn *bpf_convert_tstamp_write(const struct bpf_insn *si,
+						 struct bpf_insn *insn)
+{
+	__u8 value_reg = si->src_reg;
+	__u8 skb_reg = si->dst_reg;
+
+#ifdef CONFIG_NET_CLS_ACT
+	__u8 tmp_reg = BPF_REG_AX;
+
+	*insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg, TC_AT_INGRESS_OFFSET);
+	*insn++ = BPF_ALU32_IMM(BPF_AND, tmp_reg, TC_AT_INGRESS_MASK);
+	*insn++ = BPF_JMP32_IMM(BPF_JEQ, tmp_reg, 0, 3);
+	/* Writing __sk_buff->tstamp at ingress as the (rcv) timestamp.
+	 * Clear the skb->mono_delivery_time.
+	 */
+	*insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg,
+			      SKB_MONO_DELIVERY_TIME_OFFSET);
+	*insn++ = BPF_ALU32_IMM(BPF_AND, tmp_reg,
+				~SKB_MONO_DELIVERY_TIME_MASK);
+	*insn++ = BPF_STX_MEM(BPF_B, skb_reg, tmp_reg,
+			      SKB_MONO_DELIVERY_TIME_OFFSET);
+#endif
+
+	/* skb->tstamp = tstamp */
+	*insn++ = BPF_STX_MEM(BPF_DW, skb_reg, value_reg,
+			      offsetof(struct sk_buff, tstamp));
+	return insn;
+}
+
 static u32 bpf_convert_ctx_access(enum bpf_access_type type,
 				  const struct bpf_insn *si,
 				  struct bpf_insn *insn_buf,
@@ -9167,17 +9226,9 @@ static u32 bpf_convert_ctx_access(enum bpf_access_type type,
 		BUILD_BUG_ON(sizeof_field(struct sk_buff, tstamp) != 8);
 
 		if (type == BPF_WRITE)
-			*insn++ = BPF_STX_MEM(BPF_DW,
-					      si->dst_reg, si->src_reg,
-					      bpf_target_off(struct sk_buff,
-							     tstamp, 8,
-							     target_size));
+			insn = bpf_convert_tstamp_write(si, insn);
 		else
-			*insn++ = BPF_LDX_MEM(BPF_DW,
-					      si->dst_reg, si->src_reg,
-					      bpf_target_off(struct sk_buff,
-							     tstamp, 8,
-							     target_size));
+			insn = bpf_convert_tstamp_read(si, insn);
 		break;
 
 	case offsetof(struct __sk_buff, gso_segs):
diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c
index a77d8908e737..fea2d78b9ddc 100644
--- a/net/sched/act_bpf.c
+++ b/net/sched/act_bpf.c
@@ -53,6 +53,8 @@ static int tcf_bpf_act(struct sk_buff *skb, const struct tc_action *act,
 		bpf_compute_data_pointers(skb);
 		filter_res = bpf_prog_run(filter, skb);
 	}
+	if (unlikely(!skb->tstamp && skb->mono_delivery_time))
+		skb->mono_delivery_time = 0;
 	if (skb_sk_is_prefetched(skb) && filter_res != TC_ACT_OK)
 		skb_orphan(skb);
 
diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c
index df19a847829e..c85b85a192bf 100644
--- a/net/sched/cls_bpf.c
+++ b/net/sched/cls_bpf.c
@@ -102,6 +102,8 @@ static int cls_bpf_classify(struct sk_buff *skb, const struct tcf_proto *tp,
 			bpf_compute_data_pointers(skb);
 			filter_res = bpf_prog_run(prog->filter, skb);
 		}
+		if (unlikely(!skb->tstamp && skb->mono_delivery_time))
+			skb->mono_delivery_time = 0;
 
 		if (prog->exts_integrated) {
 			res->class   = 0;
-- 
2.30.2


  parent reply	other threads:[~2022-03-02 19:56 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-03-02 19:55 [PATCH v6 net-next 0/13] Preserve mono delivery time (EDT) in skb->tstamp Martin KaFai Lau
2022-03-02 19:55 ` [PATCH v6 net-next 01/13] net: Add skb->mono_delivery_time to distinguish mono delivery_time from (rcv) timestamp Martin KaFai Lau
2022-03-02 19:55 ` [PATCH v6 net-next 02/13] net: Add skb_clear_tstamp() to keep the mono delivery_time Martin KaFai Lau
2022-03-02 19:55 ` [PATCH v6 net-next 03/13] net: Handle delivery_time in skb->tstamp during network tapping with af_packet Martin KaFai Lau
2022-03-03 10:48   ` Daniel Borkmann
2022-03-03 19:17     ` Martin KaFai Lau
2022-03-02 19:55 ` [PATCH v6 net-next 04/13] net: Clear mono_delivery_time bit in __skb_tstamp_tx() Martin KaFai Lau
2022-03-02 19:55 ` [PATCH v6 net-next 05/13] net: Set skb->mono_delivery_time and clear it after sch_handle_ingress() Martin KaFai Lau
2022-03-02 19:55 ` [PATCH v6 net-next 06/13] net: ip: Handle delivery_time in ip defrag Martin KaFai Lau
2022-03-02 19:56 ` [PATCH v6 net-next 07/13] net: ipv6: Handle delivery_time in ipv6 defrag Martin KaFai Lau
2022-03-02 19:56 ` [PATCH v6 net-next 08/13] net: ipv6: Get rcv timestamp if needed when handling hop-by-hop IOAM option Martin KaFai Lau
2022-03-02 19:56 ` [PATCH v6 net-next 09/13] net: Get rcv tstamp if needed in nfnetlink_{log, queue}.c Martin KaFai Lau
2022-03-02 19:56 ` [PATCH v6 net-next 10/13] net: Postpone skb_clear_delivery_time() until knowing the skb is delivered locally Martin KaFai Lau
2022-03-02 20:30   ` Eric Dumazet
2022-03-02 22:33     ` Martin KaFai Lau
2022-03-02 23:41       ` Eric Dumazet
2022-03-03  0:19         ` Martin KaFai Lau
2022-03-03  0:47           ` Eric Dumazet
2022-03-02 19:56 ` Martin KaFai Lau [this message]
2022-03-03 13:00   ` [PATCH v6 net-next 11/13] bpf: Keep the (rcv) timestamp behavior for the existing tc-bpf@ingress Daniel Borkmann
2022-03-03 20:43     ` Martin KaFai Lau
2022-03-03 22:55       ` Daniel Borkmann
2022-03-03 23:42         ` Martin KaFai Lau
2022-03-02 19:56 ` [PATCH v6 net-next 12/13] bpf: Add __sk_buff->delivery_time_type and bpf_skb_set_skb_delivery_time() Martin KaFai Lau
2022-03-02 19:56 ` [PATCH v6 net-next 13/13] bpf: selftests: test skb->tstamp in redirect_neigh Martin KaFai Lau
2022-03-03  0:36 ` [PATCH v6 net-next 0/13] Preserve mono delivery time (EDT) in skb->tstamp Eric Dumazet
2022-03-03 14:50 ` patchwork-bot+netdevbpf

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220302195628.3484598-1-kafai@fb.com \
    --to=kafai@fb.com \
    --cc=andrii@kernel.org \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=kernel-team@fb.com \
    --cc=kuba@kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=willemb@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox