Received-SPF: pass (google.com: domain of linux-kernel+bounces-143845-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
Date: Sat, 13 Apr 2024 14:54:40 -0400
From: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
To: Abhishek Chauhan <quic_abchauha@quicinc.com>, 
 "David S. Miller" <davem@davemloft.net>, 
 Eric Dumazet <edumazet@google.com>, 
 Jakub Kicinski <kuba@kernel.org>, 
 Paolo Abeni <pabeni@redhat.com>, 
 netdev@vger.kernel.org, 
 linux-kernel@vger.kernel.org, 
 Andrew Halaney <ahalaney@redhat.com>, 
 Willem de Bruijn <willemdebruijn.kernel@gmail.com>, 
 Martin KaFai Lau <martin.lau@kernel.org>, 
 Martin KaFai Lau <martin.lau@linux.dev>, 
 Daniel Borkmann <daniel@iogearbox.net>, 
 bpf <bpf@vger.kernel.org>
Cc: kernel@quicinc.com
Message-ID: <661ad4f0e3766_3be9a7294a1@willemb.c.googlers.com.notmuch>
In-Reply-To: <20240412210125.1780574-2-quic_abchauha@quicinc.com>
References: <20240412210125.1780574-1-quic_abchauha@quicinc.com>
 <20240412210125.1780574-2-quic_abchauha@quicinc.com>
Subject: Re: [RFC PATCH bpf-next v3 1/2] net: Rename mono_delivery_time to
 tstamp_type for scalabilty
Precedence: bulk
Mime-Version: 1.0
Content-Type: text/plain;
 charset=utf-8
Content-Transfer-Encoding: 7bit

Abhishek Chauhan wrote:
> 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 out if enum defined as
> part of 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 enum to distinguish between mono and real time
> 
> Bridge driver today has no support to forward the userspace timestamp
> packets and ends up resetting the timestamp. ETF qdisc checks the
> packet coming from userspace and encounters to be 0 thereby dropping
> time sensitive packets. These changes will allow userspace timestamps
> packets to be forwarded from the bridge to NIC drivers.
> 
> 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@linux.dev/
> Signed-off-by: Abhishek Chauhan <quic_abchauha@quicinc.com>
> ---
> 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 
>   as mentioned by Willem.
> 
>  include/linux/skbuff.h                     | 33 +++++++++++++++-------
>  include/net/inet_frag.h                    |  4 +--
>  net/bridge/netfilter/nf_conntrack_bridge.c |  6 ++--
>  net/core/dev.c                             |  2 +-
>  net/core/filter.c                          |  8 +++---
>  net/ipv4/inet_fragment.c                   |  2 +-
>  net/ipv4/ip_fragment.c                     |  2 +-
>  net/ipv4/ip_output.c                       |  8 +++---
>  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 +--
>  16 files changed, 59 insertions(+), 46 deletions(-)
> 
> diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
> index 7135a3e94afd..a83a2120b57f 100644
> --- a/include/linux/skbuff.h
> +++ b/include/linux/skbuff.h
> @@ -702,6 +702,11 @@ typedef unsigned int sk_buff_data_t;
>  typedef unsigned char *sk_buff_data_t;
>  #endif
>  
> +enum skb_tstamp_type {
> +	CLOCK_REAL = 0, /* Time base is realtime */
> +	CLOCK_MONO = 1, /* Time base is Monotonic */
> +};

Minor: inconsistent capitalization

> +
>  /**
>   * DOC: Basic sk_buff geometry
>   *
> @@ -819,7 +824,7 @@ 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 +955,7 @@ 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_MONO_DELIVERY_TIME_MASK */

Also remove reference to MONO_DELIVERY_TIME_MASK, or instead refer to
skb_tstamp_type.

>  #ifdef CONFIG_NET_XGRESS
>  	__u8			tc_at_ingress:1;	/* See TC_AT_INGRESS_MASK */
>  	__u8			tc_skip_classify:1;
> @@ -4237,7 +4242,7 @@ static inline void skb_get_new_timestampns(const struct sk_buff *skb,
>  static inline void __net_timestamp(struct sk_buff *skb)
>  {
>  	skb->tstamp = ktime_get_real();
> -	skb->mono_delivery_time = 0;
> +	skb->tstamp_type = CLOCK_REAL;
>  }
>  
>  static inline ktime_t net_timedelta(ktime_t t)
> @@ -4246,10 +4251,18 @@ static inline ktime_t net_timedelta(ktime_t t)
>  }
>  
>  static inline void skb_set_delivery_time(struct sk_buff *skb, ktime_t kt,
> -					 bool mono)
> +					  u8 tstamp_type)
>  {
>  	skb->tstamp = kt;
> -	skb->mono_delivery_time = kt && mono;
> +
> +	switch (tstamp_type) {
> +	case CLOCK_REAL:
> +		skb->tstamp_type = CLOCK_REAL;
> +		break;
> +	case CLOCK_MONO:
> +		skb->tstamp_type = kt && tstamp_type;
> +		break;
> +	}

Technically this leaves the tstamp_type undefined if (skb, 0, CLOCK_REAL)
>  }
>  
>  DECLARE_STATIC_KEY_FALSE(netstamp_needed_key);
> @@ -4259,8 +4272,8 @@ 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 = 0;
> +	if (skb->tstamp_type) {
> +		skb->tstamp_type = CLOCK_REAL;
>  		if (static_branch_unlikely(&netstamp_needed_key))
>  			skb->tstamp = ktime_get_real();
>  		else
> @@ -4270,7 +4283,7 @@ static inline void skb_clear_delivery_time(struct sk_buff *skb)
>  
>  static inline void skb_clear_tstamp(struct sk_buff *skb)
>  {
> -	if (skb->mono_delivery_time)
> +	if (skb->tstamp_type)
>  		return;
>  
>  	skb->tstamp = 0;
> @@ -4278,7 +4291,7 @@ static inline void skb_clear_tstamp(struct sk_buff *skb)
>  
>  static inline ktime_t skb_tstamp(const struct sk_buff *skb)
>  {
> -	if (skb->mono_delivery_time)
> +	if (skb->tstamp_type == CLOCK_MONO)
>  		return 0;

Should this be if (skb->tstamp_type), in line with skb_clear_tstamp,
right above?

> @@ -1649,7 +1649,7 @@ void ip_send_unicast_reply(struct sock *sk, struct sk_buff *skb,
>  			  arg->csumoffset) = csum_fold(csum_add(nskb->csum,
>  								arg->csum));
>  		nskb->ip_summed = CHECKSUM_NONE;
> -		nskb->mono_delivery_time = !!transmit_time;
> +		nskb->tstamp_type = !!transmit_time;

In anticipation of more tstamp_types, explicitly set to CLOCK_MONO.

>  		if (txhash)
>  			skb_set_hash(nskb, txhash, PKT_HASH_TYPE_L4);
>  		ip_push_pending_frames(sk, &fl4);