Commit 15f295f5 authored by D. Wythe's avatar D. Wythe Committed by Martin KaFai Lau
Browse files

net/smc: bpf: Introduce generic hook for handshake flow 



The introduction of IPPROTO_SMC enables eBPF programs to determine
whether to use SMC based on the context of socket creation, such as
network namespaces, PID and comm name, etc.

As a subsequent enhancement, to introduce a new generic hook that
allows decisions on whether to use SMC or not at runtime, including
but not limited to local/remote IP address or ports.

User can write their own implememtion via bpf_struct_ops now to choose
whether to use SMC or not before TCP 3rd handshake to be comleted.

Signed-off-by: default avatarD. Wythe <alibuda@linux.alibaba.com>
Signed-off-by: default avatarMartin KaFai Lau <martin.lau@kernel.org>
Reviewed-by: default avatarDust Li <dust.li@linux.alibaba.com>
Link: https://patch.msgid.link/20251107035632.115950-3-alibuda@linux.alibaba.com
parent 07c428ec

include/net/netns/smc.h

+3 −0
Original line number Diff line number Diff line
@@ -17,6 +17,9 @@ struct netns_smc { 
#ifdef CONFIG_SYSCTL

	struct ctl_table_header		*smc_hdr;

#endif

#if IS_ENABLED(CONFIG_SMC_HS_CTRL_BPF)

	struct smc_hs_ctrl __rcu	*hs_ctrl;

#endif /* CONFIG_SMC_HS_CTRL_BPF */

	unsigned int			sysctl_autocorking_size;

	unsigned int			sysctl_smcr_buf_type;

	int				sysctl_smcr_testlink_time;

include/net/smc.h

+53 −0
Original line number Diff line number Diff line
@@ -17,6 +17,8 @@ 
#include <linux/wait.h>

#include <linux/dibs.h>



struct tcp_sock;

struct inet_request_sock;

struct sock;



#define SMC_MAX_PNETID_LEN	16	/* Max. length of PNET id */
@@ -50,4 +52,55 @@ struct smcd_dev { 
	u8 going_away : 1;

};



#define SMC_HS_CTRL_NAME_MAX 16



enum {

	/* ops can be inherit from init_net */

	SMC_HS_CTRL_FLAG_INHERITABLE = 0x1,



	SMC_HS_CTRL_ALL_FLAGS = SMC_HS_CTRL_FLAG_INHERITABLE,

};



struct smc_hs_ctrl {

	/* private */



	struct list_head list;

	struct module *owner;



	/* public */



	/* unique name */

	char name[SMC_HS_CTRL_NAME_MAX];

	int flags;



	/* Invoked before computing SMC option for SYN packets.

	 * We can control whether to set SMC options by returning various value.

	 * Return 0 to disable SMC, or return any other value to enable it.

	 */

	int (*syn_option)(struct tcp_sock *tp);



	/* Invoked before Set up SMC options for SYN-ACK packets

	 * We can control whether to respond SMC options by returning various

	 * value. Return 0 to disable SMC, or return any other value to enable

	 * it.

	 */

	int (*synack_option)(const struct tcp_sock *tp,

			     struct inet_request_sock *ireq);

};



#if IS_ENABLED(CONFIG_SMC_HS_CTRL_BPF)

#define smc_call_hsbpf(init_val, tp, func, ...) ({				\

	typeof(init_val) __ret = (init_val);					\

	struct smc_hs_ctrl *ctrl;						\

	rcu_read_lock();							\

	ctrl = rcu_dereference(sock_net((struct sock *)(tp))->smc.hs_ctrl);	\

	if (ctrl && ctrl->func)							\

		__ret = ctrl->func(tp, ##__VA_ARGS__);				\

	rcu_read_unlock();							\

	__ret;									\

})

#else

#define smc_call_hsbpf(init_val, tp, ...)  ({ (void)(tp); (init_val); })

#endif /* CONFIG_SMC_HS_CTRL_BPF */



#endif	/* _SMC_H */

net/ipv4/tcp_output.c

+17 −14
Original line number Diff line number Diff line
@@ -40,6 +40,7 @@ 
#include <net/tcp.h>

#include <net/tcp_ecn.h>

#include <net/mptcp.h>

#include <net/smc.h>

#include <net/proto_memory.h>

#include <net/psp.h>
@@ -802,36 +803,38 @@ static void tcp_options_write(struct tcphdr *th, struct tcp_sock *tp, 
	mptcp_options_write(th, ptr, tp, opts);

}



static void smc_set_option(const struct tcp_sock *tp,

static void smc_set_option(struct tcp_sock *tp,

			   struct tcp_out_options *opts,

			   unsigned int *remaining)

{

#if IS_ENABLED(CONFIG_SMC)

	if (static_branch_unlikely(&tcp_have_smc)) {

		if (tp->syn_smc) {

			if (*remaining >= TCPOLEN_EXP_SMC_BASE_ALIGNED) {

	if (static_branch_unlikely(&tcp_have_smc) && tp->syn_smc) {

		tp->syn_smc = !!smc_call_hsbpf(1, tp, syn_option);

		/* re-check syn_smc */

		if (tp->syn_smc &&

		    *remaining >= TCPOLEN_EXP_SMC_BASE_ALIGNED) {

			opts->options |= OPTION_SMC;

			*remaining -= TCPOLEN_EXP_SMC_BASE_ALIGNED;

		}

	}

	}

#endif

}



static void smc_set_option_cond(const struct tcp_sock *tp,

				const struct inet_request_sock *ireq,

				struct inet_request_sock *ireq,

				struct tcp_out_options *opts,

				unsigned int *remaining)

{

#if IS_ENABLED(CONFIG_SMC)

	if (static_branch_unlikely(&tcp_have_smc)) {

		if (tp->syn_smc && ireq->smc_ok) {

			if (*remaining >= TCPOLEN_EXP_SMC_BASE_ALIGNED) {

	if (static_branch_unlikely(&tcp_have_smc) && tp->syn_smc && ireq->smc_ok) {

		ireq->smc_ok = !!smc_call_hsbpf(1, tp, synack_option, ireq);

		/* re-check smc_ok */

		if (ireq->smc_ok &&

		    *remaining >= TCPOLEN_EXP_SMC_BASE_ALIGNED) {

			opts->options |= OPTION_SMC;

			*remaining -= TCPOLEN_EXP_SMC_BASE_ALIGNED;

		}

	}

	}

#endif

}

net/smc/Kconfig

+10 −0
Original line number Diff line number Diff line
@@ -19,3 +19,13 @@ config SMC_DIAG 
	  smcss.



	  if unsure, say Y.



config SMC_HS_CTRL_BPF

	bool "Generic eBPF hook for SMC handshake flow"

	depends on SMC && BPF_SYSCALL

	default y

	help

	  SMC_HS_CTRL_BPF enables support to register generic eBPF hook for SMC

	  handshake flow, which offer much greater flexibility in modifying the behavior

	  of the SMC protocol stack compared to a complete kernel-based approach. Select

	  this option if you want filtring the handshake process via eBPF programs.
 
 No newline at end of file

net/smc/Makefile

+1 −0
Original line number Diff line number Diff line
@@ -6,3 +6,4 @@ smc-y := af_smc.o smc_pnet.o smc_ib.o smc_clc.o smc_core.o smc_wr.o smc_llc.o 
smc-y += smc_cdc.o smc_tx.o smc_rx.o smc_close.o smc_ism.o smc_netlink.o smc_stats.o

smc-y += smc_tracepoint.o smc_inet.o

smc-$(CONFIG_SYSCTL) += smc_sysctl.o

smc-$(CONFIG_SMC_HS_CTRL_BPF) += smc_hs_bpf.o