Commit 1693c5db authored by Abhishek Chauhan's avatar Abhishek Chauhan Committed by Martin KaFai Lau
Browse files

net: Add additional bit to support clockid_t timestamp type 



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 existing support CLOCK_MONOTONIC and CLOCK_REALTIME.

Signed-off-by: default avatarAbhishek Chauhan <quic_abchauha@quicinc.com>
Reviewed-by: default avatarWillem de Bruijn <willemb@google.com>
Reviewed-by: default avatarMartin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20240509211834.3235191-3-quic_abchauha@quicinc.com


Signed-off-by: default avatarMartin KaFai Lau <martin.lau@kernel.org>
parent 4d25ca2d

include/linux/skbuff.h

+12 −6
Original line number Diff line number Diff line
@@ -709,6 +709,8 @@ typedef unsigned char *sk_buff_data_t; 
enum skb_tstamp_type {

	SKB_CLOCK_REALTIME,

	SKB_CLOCK_MONOTONIC,

	SKB_CLOCK_TAI,

	__SKB_CLOCK_MAX = SKB_CLOCK_TAI,

};



/**
@@ -957,7 +959,7 @@ struct sk_buff { 
	/* private: */

	__u8			__mono_tc_offset[0];

	/* public: */

	__u8			tstamp_type:1;	/* See skb_tstamp_type */

	__u8			tstamp_type:2;	/* See skb_tstamp_type */

#ifdef CONFIG_NET_XGRESS

	__u8			tc_at_ingress:1;	/* See TC_AT_INGRESS_MASK */

	__u8			tc_skip_classify:1;
@@ -1087,15 +1089,16 @@ struct sk_buff { 
#endif

#define PKT_TYPE_OFFSET		offsetof(struct sk_buff, __pkt_type_offset)



/* if you move tc_at_ingress or mono_delivery_time

/* if you move tc_at_ingress or tstamp_type

 * around, you also must adapt these constants.

 */

#ifdef __BIG_ENDIAN_BITFIELD

#define SKB_MONO_DELIVERY_TIME_MASK	(1 << 7)

#define TC_AT_INGRESS_MASK		(1 << 6)

#define SKB_TSTAMP_TYPE_MASK		(3 << 6)

#define SKB_TSTAMP_TYPE_RSHIFT		(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_TSTAMP_TYPE_MASK		(3)

#define TC_AT_INGRESS_MASK		(1 << 2)

#endif

#define SKB_BF_MONO_TC_OFFSET		offsetof(struct sk_buff, __mono_tc_offset)
@@ -4216,6 +4219,9 @@ static inline void skb_set_delivery_type_by_clockid(struct sk_buff *skb, 
	case CLOCK_MONOTONIC:

		tstamp_type = SKB_CLOCK_MONOTONIC;

		break;

	case CLOCK_TAI:

		tstamp_type = SKB_CLOCK_TAI;

		break;

	default:

		WARN_ON_ONCE(1);

		kt = 0;

include/uapi/linux/bpf.h

+10 −5
Original line number Diff line number Diff line
@@ -6207,12 +6207,17 @@ union { \ 
	__u64 :64;			\

} __attribute__((aligned(8)))



/* The enum used in skb->tstamp_type. It specifies the clock type

 * of the time stored in the skb->tstamp.

 */

enum {

	BPF_SKB_TSTAMP_UNSPEC,

	BPF_SKB_TSTAMP_DELIVERY_MONO,	/* tstamp has mono 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.

	BPF_SKB_TSTAMP_UNSPEC = 0,		/* DEPRECATED */

	BPF_SKB_TSTAMP_DELIVERY_MONO = 1,	/* DEPRECATED */

	BPF_SKB_CLOCK_REALTIME = 0,

	BPF_SKB_CLOCK_MONOTONIC = 1,

	BPF_SKB_CLOCK_TAI = 2,

	/* For any future BPF_SKB_CLOCK_* that the bpf prog cannot handle,

	 * the bpf prog can try to deduce it by ingress/egress/skb->sk->sk_clockid.

	 */

};

net/core/filter.c

+26 −20
Original line number Diff line number Diff line
@@ -7726,17 +7726,21 @@ BPF_CALL_3(bpf_skb_set_tstamp, struct sk_buff *, skb, 
		return -EOPNOTSUPP;



	switch (tstamp_type) {

	case BPF_SKB_TSTAMP_DELIVERY_MONO:

	case BPF_SKB_CLOCK_REALTIME:

		skb->tstamp = tstamp;

		skb->tstamp_type = SKB_CLOCK_REALTIME;

		break;

	case BPF_SKB_CLOCK_MONOTONIC:

		if (!tstamp)

			return -EINVAL;

		skb->tstamp = tstamp;

		skb->tstamp_type = SKB_CLOCK_MONOTONIC;

		break;

	case BPF_SKB_TSTAMP_UNSPEC:

		if (tstamp)

	case BPF_SKB_CLOCK_TAI:

		if (!tstamp)

			return -EINVAL;

		skb->tstamp = 0;

		skb->tstamp_type = SKB_CLOCK_REALTIME;

		skb->tstamp = tstamp;

		skb->tstamp_type = SKB_CLOCK_TAI;

		break;

	default:

		return -EINVAL;
@@ -9387,16 +9391,17 @@ static struct bpf_insn *bpf_convert_tstamp_type_read(const struct bpf_insn *si, 
{

	__u8 value_reg = si->dst_reg;

	__u8 skb_reg = si->src_reg;

	/* AX is needed because src_reg and dst_reg could be the same */

	__u8 tmp_reg = BPF_REG_AX;



	*insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg,

			      SKB_BF_MONO_TC_OFFSET);

	*insn++ = BPF_JMP32_IMM(BPF_JSET, tmp_reg,

				SKB_MONO_DELIVERY_TIME_MASK, 2);

	*insn++ = BPF_MOV32_IMM(value_reg, BPF_SKB_TSTAMP_UNSPEC);

	*insn++ = BPF_JMP_A(1);

	*insn++ = BPF_MOV32_IMM(value_reg, BPF_SKB_TSTAMP_DELIVERY_MONO);

	BUILD_BUG_ON(__SKB_CLOCK_MAX != (int)BPF_SKB_CLOCK_TAI);

	BUILD_BUG_ON(SKB_CLOCK_REALTIME != (int)BPF_SKB_CLOCK_REALTIME);

	BUILD_BUG_ON(SKB_CLOCK_MONOTONIC != (int)BPF_SKB_CLOCK_MONOTONIC);

	BUILD_BUG_ON(SKB_CLOCK_TAI != (int)BPF_SKB_CLOCK_TAI);

	*insn++ = BPF_LDX_MEM(BPF_B, value_reg, skb_reg, SKB_BF_MONO_TC_OFFSET);

	*insn++ = BPF_ALU32_IMM(BPF_AND, value_reg, SKB_TSTAMP_TYPE_MASK);

#ifdef __BIG_ENDIAN_BITFIELD

	*insn++ = BPF_ALU32_IMM(BPF_RSH, value_reg, SKB_TSTAMP_TYPE_RSHIFT);

#else

	BUILD_BUG_ON(!(SKB_TSTAMP_TYPE_MASK & 0x1));

#endif



	return insn;

}
@@ -9439,10 +9444,11 @@ static struct bpf_insn *bpf_convert_tstamp_read(const struct bpf_prog *prog, 
		__u8 tmp_reg = BPF_REG_AX;



		*insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg, SKB_BF_MONO_TC_OFFSET);

		*insn++ = BPF_ALU32_IMM(BPF_AND, tmp_reg,

					TC_AT_INGRESS_MASK | SKB_MONO_DELIVERY_TIME_MASK);

		*insn++ = BPF_JMP32_IMM(BPF_JNE, tmp_reg,

					TC_AT_INGRESS_MASK | SKB_MONO_DELIVERY_TIME_MASK, 2);

		/* check if ingress mask bits is set */

		*insn++ = BPF_JMP32_IMM(BPF_JSET, tmp_reg, TC_AT_INGRESS_MASK, 1);

		*insn++ = BPF_JMP_A(4);

		*insn++ = BPF_JMP32_IMM(BPF_JSET, tmp_reg, SKB_TSTAMP_TYPE_MASK, 1);

		*insn++ = BPF_JMP_A(2);

		/* skb->tc_at_ingress && skb->tstamp_type,

		 * read 0 as the (rcv) timestamp.

		 */
@@ -9479,7 +9485,7 @@ static struct bpf_insn *bpf_convert_tstamp_write(const struct bpf_prog *prog, 
		/* goto <store> */

		*insn++ = BPF_JMP_A(2);

		/* <clear>: skb->tstamp_type */

		*insn++ = BPF_ALU32_IMM(BPF_AND, tmp_reg, ~SKB_MONO_DELIVERY_TIME_MASK);

		*insn++ = BPF_ALU32_IMM(BPF_AND, tmp_reg, ~SKB_TSTAMP_TYPE_MASK);

		*insn++ = BPF_STX_MEM(BPF_B, skb_reg, tmp_reg, SKB_BF_MONO_TC_OFFSET);

	}

#endif

net/ipv4/ip_output.c

+4 −1
Original line number Diff line number Diff line
@@ -1457,7 +1457,10 @@ struct sk_buff *__ip_make_skb(struct sock *sk, 


	skb->priority = (cork->tos != -1) ? cork->priority: READ_ONCE(sk->sk_priority);

	skb->mark = cork->mark;

	skb->tstamp = cork->transmit_time;

	if (sk_is_tcp(sk))

		skb_set_delivery_time(skb, cork->transmit_time, SKB_CLOCK_MONOTONIC);

	else

		skb_set_delivery_type_by_clockid(skb, cork->transmit_time, sk->sk_clockid);

	/*

	 * Steal rt from cork.dst to avoid a pair of atomic_inc/atomic_dec

	 * on dst refcount

net/ipv4/raw.c

+1 −1
Original line number Diff line number Diff line
@@ -360,7 +360,7 @@ static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4, 
	skb->protocol = htons(ETH_P_IP);

	skb->priority = READ_ONCE(sk->sk_priority);

	skb->mark = sockc->mark;

	skb->tstamp = sockc->transmit_time;

	skb_set_delivery_type_by_clockid(skb, sockc->transmit_time, sk->sk_clockid);

	skb_dst_set(skb, &rt->dst);

	*rtp = NULL;