History

bpf: Add __sk_buff->delivery_time_type and bpf_skb_set_skb_delivery_time()

* __sk_buff->delivery_time_type: This patch adds __sk_buff->delivery_time_type. It tells if the delivery_time is stored in __sk_buff->tstamp or not. It will be most useful for ingress to tell if the __sk_buff->tstamp has the (rcv) timestamp or delivery_time. If delivery_time_type is 0 (BPF_SKB_DELIVERY_TIME_NONE), it has the (rcv) timestamp. Two non-zero types are defined for the delivery_time_type, BPF_SKB_DELIVERY_TIME_MONO and BPF_SKB_DELIVERY_TIME_UNSPEC. For UNSPEC, it can only happen in egress because only mono delivery_time can be forwarded to ingress now. The clock of UNSPEC delivery_time can be deduced from the skb->sk->sk_clockid which is how the sch_etf doing it also. * Provide forwarded delivery_time to tc-bpf@ingress: With the help of the new delivery_time_type, the tc-bpf has a way to tell if the __sk_buff->tstamp has the (rcv) timestamp or the delivery_time. During bpf load time, the verifier will learn if the bpf prog has accessed the new __sk_buff->delivery_time_type. If it does, it means the tc-bpf@ingress is expecting the skb->tstamp could have the delivery_time. The kernel will then read the skb->tstamp as-is during bpf insn rewrite without checking the skb->mono_delivery_time. This is done by adding a new prog->delivery_time_access bit. The same goes for writing skb->tstamp. * bpf_skb_set_delivery_time(): The bpf_skb_set_delivery_time() helper is added to allow setting both delivery_time and the delivery_time_type at the same time. If the tc-bpf does not need to change the delivery_time_type, it can directly write to the __sk_buff->tstamp as the existing tc-bpf has already been doing. It will be most useful at ingress to change the __sk_buff->tstamp from the (rcv) timestamp to a mono delivery_time and then bpf_redirect_*(). bpf only has mono clock helper (bpf_ktime_get_ns), and the current known use case is the mono EDT for fq, and only mono delivery time can be kept during forward now, so bpf_skb_set_delivery_time() only supports setting BPF_SKB_DELIVERY_TIME_MONO. It can be extended later when use cases come up and the forwarding path also supports other clock bases.

Martin KaFai Lau committed 4 years ago

Commit 8d21ec01 parent 74491974 files · +216 −38Browse files

Showing the changes to include/uapi (1 of 4 files). Show all

Unified · Split

include/uapi/linux/bpf.h+40−1
@@ -5086,6 +5086,37 @@ union bpf_attr { · 5085 unmodified lines
50865086 * Return
50875087 * 0 on success, or a negative error in case of failure. On error
50885088 * *dst* buffer is zeroed out.
5089+ *
5090+ * long bpf_skb_set_delivery_time(struct sk_buff *skb, u64 dtime, u32 dtime_type)
5091+ * Description
5092+ * Set a *dtime* (delivery time) to the __sk_buff->tstamp and also
5093+ * change the __sk_buff->delivery_time_type to *dtime_type*.
5094+ *
5095+ * When setting a delivery time (non zero *dtime*) to
5096+ * __sk_buff->tstamp, only BPF_SKB_DELIVERY_TIME_MONO *dtime_type*
5097+ * is supported. It is the only delivery_time_type that will be
5098+ * kept after bpf_redirect_*().
5099+ *
5100+ * If there is no need to change the __sk_buff->delivery_time_type,
5101+ * the delivery time can be directly written to __sk_buff->tstamp
5102+ * instead.
5103+ *
5104+ * *dtime* 0 and *dtime_type* BPF_SKB_DELIVERY_TIME_NONE
5105+ * can be used to clear any delivery time stored in
5106+ * __sk_buff->tstamp.
5107+ *
5108+ * Only IPv4 and IPv6 skb->protocol are supported.
5109+ *
5110+ * This function is most useful when it needs to set a
5111+ * mono delivery time to __sk_buff->tstamp and then
5112+ * bpf_redirect_*() to the egress of an iface. For example,
5113+ * changing the (rcv) timestamp in __sk_buff->tstamp at
5114+ * ingress to a mono delivery time and then bpf_redirect_*()
5115+ * to sch_fq@phy-dev.
5116+ * Return
5117+ * 0 on success.
5118+ * **-EINVAL** for invalid input
5119+ * **-EOPNOTSUPP** for unsupported delivery_time_type and protocol
50895120 */
50905121 #define __BPF_FUNC_MAPPER(FN) \
50915122 FN(unspec), \
@@ -5280,6 +5311,7 @@ union bpf_attr { · 188 unmodified lines
52805311 FN(xdp_load_bytes), \
52815312 FN(xdp_store_bytes), \
52825313 FN(copy_from_user_task), \
5314+ FN(skb_set_delivery_time), \
52835315 /* */
52845316
52855317 /* integer value in 'imm' field of BPF_CALL instruction selects which helper
@@ -5469,6 +5501,12 @@ union { \ · 183 unmodified lines
54695501 __u64 :64; \
54705502 } __attribute__((aligned(8)))
54715503
5504+enum {
5505+ BPF_SKB_DELIVERY_TIME_NONE,
5506+ BPF_SKB_DELIVERY_TIME_UNSPEC,
5507+ BPF_SKB_DELIVERY_TIME_MONO,
5508+};
5509+
54725510 /* user accessible mirror of in-kernel sk_buff.
54735511 * new fields can only be added to the end of this structure
54745512 */
@@ -5509,7 +5547,8 @@ struct __sk_buff { · 34 unmodified lines
55095547 __u32 gso_segs;
55105548 __bpf_md_ptr(struct bpf_sock *, sk);
55115549 __u32 gso_size;
5512− __u32 :32; /* Padding, future use. */
5550+ __u8 delivery_time_type;
5551+ __u32 :24; /* Padding, future use. */
55135552 __u64 hwtstamp;
55145553 };
55155554