Date: Fri, 12 May 2023 19:36:32 -0700
From: Yonghong Song <yhs@...a.com>
To: "D. Wythe" <alibuda@...ux.alibaba.com>, kgraul@...ux.ibm.com,
        wenjia@...ux.ibm.com, jaka@...ux.ibm.com, ast@...nel.org,
        daniel@...earbox.net, andrii@...nel.org, martin.lau@...ux.dev,
        pabeni@...hat.com, song@...nel.org, sdf@...gle.com, haoluo@...gle.com,
        yhs@...com, edumazet@...gle.com, john.fastabend@...il.com,
        kpsingh@...nel.org, jolsa@...nel.org, guwen@...ux.alibaba.com
Cc: kuba@...nel.org, davem@...emloft.net, netdev@...r.kernel.org,
        linux-s390@...r.kernel.org, linux-rdma@...r.kernel.org,
        bpf@...r.kernel.org
Subject: Re: [PATCH bpf-next v1 4/5] bpf: add smc negotiator support in BPF
 struct_ops



On 5/11/23 11:24 PM, D. Wythe wrote:
> From: "D. Wythe" <alibuda@...ux.alibaba.com>
> 
> This PATCH attempts to introduce BPF injection capability for SMC.
> Considering that the SMC protocol is not suitable for all scenarios,
> especially for short-lived. However, for most applications, they cannot
> guarantee that there are no such scenarios at all. Therefore, apps
> may need some specific strategies to decide shall we need to use SMC
> or not, for example, apps can limit the scope of the SMC to a specific
> IP address or port.
> 
> Based on the consideration of transparent replacement, we hope that apps
> can remain transparent even if they need to formulate some specific
> strategies for SMC using. That is, do not need to recompile their code.
> 
> On the other hand, we need to ensure the scalability of strategies
> implementation. Although it is simple to use socket options or sysctl,
> it will bring more complexity to subsequent expansion.
> 
> Fortunately, BPF can solve these concerns very well, users can write
> thire own strategies in eBPF to choose whether to use SMC or not.
> And it's quite easy for them to modify their strategies in the future.
> 
> This PATCH implement injection capability for SMC via struct_ops.
> In that way, we can add new injection scenarios in the future.
> 
> Signed-off-by: D. Wythe <alibuda@...ux.alibaba.com>
> ---
>   kernel/bpf/bpf_struct_ops_types.h |   4 +
>   net/Makefile                      |   2 +-
>   net/smc/bpf_smc.c                 | 171 ++++++++++++++++++++++++++++++++++++++
>   3 files changed, 176 insertions(+), 1 deletion(-)
>   create mode 100644 net/smc/bpf_smc.c
> 
> diff --git a/kernel/bpf/bpf_struct_ops_types.h b/kernel/bpf/bpf_struct_ops_types.h
> index 5678a9d..d952b85 100644
> --- a/kernel/bpf/bpf_struct_ops_types.h
> +++ b/kernel/bpf/bpf_struct_ops_types.h
> @@ -9,4 +9,8 @@
>   #include <net/tcp.h>
>   BPF_STRUCT_OPS_TYPE(tcp_congestion_ops)
>   #endif
> +#if IS_ENABLED(CONFIG_SMC_BPF)
> +#include <net/smc.h>
> +BPF_STRUCT_OPS_TYPE(smc_sock_negotiator_ops)
> +#endif
>   #endif
> diff --git a/net/Makefile b/net/Makefile
> index 222916a..2139fa4 100644
> --- a/net/Makefile
> +++ b/net/Makefile
> @@ -52,7 +52,7 @@ obj-$(CONFIG_TIPC)		+= tipc/
>   obj-$(CONFIG_NETLABEL)		+= netlabel/
>   obj-$(CONFIG_IUCV)		+= iucv/
>   obj-$(CONFIG_SMC)		+= smc/
> -obj-$(CONFIG_SMC_BPF)		+= smc/smc_negotiator.o
> +obj-$(CONFIG_SMC_BPF)		+= smc/smc_negotiator.o smc/bpf_smc.o
>   obj-$(CONFIG_RFKILL)		+= rfkill/
>   obj-$(CONFIG_NET_9P)		+= 9p/
>   obj-$(CONFIG_CAIF)		+= caif/
> diff --git a/net/smc/bpf_smc.c b/net/smc/bpf_smc.c
> new file mode 100644
> index 0000000..ac9a9ae91
> --- /dev/null
> +++ b/net/smc/bpf_smc.c
> @@ -0,0 +1,171 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +/*
> + *  Support eBPF for Shared Memory Communications over RDMA (SMC-R) and RoCE
> + *
> + *  Copyright IBM Corp. 2016, 2018

The above description and copyright sound very wierd.

> + *
> + *  Author(s):  D. Wythe <alibuda@...ux.alibaba.com>

One author, so just "Author: ...".
> + */
> +
> +#include <linux/bpf_verifier.h>
> +#include <linux/btf_ids.h>
> +#include <linux/kernel.h>
> +#include <linux/bpf.h>
> +#include <linux/btf.h>
> +#include "smc_negotiator.h"
> +
> +extern struct bpf_struct_ops bpf_smc_sock_negotiator_ops;
> +static u32 smc_sock_id, sock_id;
> +
> +static int bpf_smc_negotiator_init(struct btf *btf)
> +{
> +	s32 type_id;
> +
> +	type_id = btf_find_by_name_kind(btf, "sock", BTF_KIND_STRUCT);
> +	if (type_id < 0)
> +		return -EINVAL;
> +	sock_id = type_id;
> +
> +	type_id = btf_find_by_name_kind(btf, "smc_sock", BTF_KIND_STRUCT);
> +	if (type_id < 0)
> +		return -EINVAL;
> +	smc_sock_id = type_id;
> +
> +	return 0;
> +}
> +
> +/* register ops */
> +static int bpf_smc_negotiator_reg(void *kdata)
> +{
> +	return smc_sock_register_negotiator_ops(kdata);
> +}
> +
> +/* unregister ops */
> +static void bpf_smc_negotiator_unreg(void *kdata)
> +{
> +	smc_sock_unregister_negotiator_ops(kdata);
> +}
> +
> +/* unregister ops */

update ops?
Also I think the above comments like
'register ops', 'unregister ops' and 'update ops' are not
necessary. The code itself is self-explanary.

> +static int bpf_smc_negotiator_update(void *kdata, void *old_kdata)
> +{
> +	return smc_sock_update_negotiator_ops(kdata, old_kdata);
> +}
> +
> +static int bpf_smc_negotiator_validate(void *kdata)
> +{
> +	return smc_sock_validate_negotiator_ops(kdata);
> +}
> +
> +static int bpf_smc_negotiator_check_member(const struct btf_type *t,
> +					   const struct btf_member *member,
> +					   const struct bpf_prog *prog)
> +{
> +	return 0;
> +}
> +
> +static int bpf_smc_negotiator_init_member(const struct btf_type *t,
> +					  const struct btf_member *member,
> +					  void *kdata, const void *udata)
> +{
> +	const struct smc_sock_negotiator_ops *uops;
> +	struct smc_sock_negotiator_ops *ops;
> +	u32 moff;
> +
> +	uops = (const struct smc_sock_negotiator_ops *)udata;
> +	ops = (struct smc_sock_negotiator_ops *)kdata;
> +
> +	moff = __btf_member_bit_offset(t, member) / 8;
> +
> +	/* init name */
> +	if (moff ==  offsetof(struct smc_sock_negotiator_ops, name)) {
> +		if (bpf_obj_name_cpy(ops->name, uops->name,
> +				     sizeof(uops->name)) <= 0)
> +			return -EINVAL;
> +		return 1;
> +	}
> +
> +	return 0;
> +}
> +
> +BPF_CALL_1(bpf_smc_skc_to_tcp_sock, struct sock *, sk)
> +{
> +	if (sk && sk_fullsock(sk) && sk->sk_family == AF_SMC)
> +		return (unsigned long)((struct smc_sock *)(sk))->clcsock->sk;
> +
> +	return (unsigned long)NULL;
> +}
> +
> +static const struct bpf_func_proto bpf_smc_skc_to_tcp_sock_proto = {
> +	.func			= bpf_smc_skc_to_tcp_sock,
> +	.gpl_only		= false,
> +	.ret_type		= RET_PTR_TO_BTF_ID_OR_NULL,
> +	.arg1_type		= ARG_PTR_TO_BTF_ID_SOCK_COMMON,
> +	.ret_btf_id		= &btf_sock_ids[BTF_SOCK_TYPE_TCP],
> +};
> +
> +static const struct bpf_func_proto *
> +smc_negotiator_prog_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
> +{
> +	const struct btf_member *m;
> +	const struct btf_type *t;
> +	u32 midx, moff;
> +
> +	midx = prog->expected_attach_type;
> +	t = bpf_smc_sock_negotiator_ops.type;
> +	m = &btf_type_member(t)[midx];
> +
> +	moff = __btf_member_bit_offset(t, m) / 8;
> +
> +	switch (func_id) {
> +	case BPF_FUNC_setsockopt:
> +		switch (moff) {
> +		/* Avoid potential deadloop risk */
> +		case offsetof(struct smc_sock_negotiator_ops, init):
> +			fallthrough;

I am not sure whether a 'fallthrough' is needed here or since the case
itself does not have any code. Any warning will show up if
'fallthrough;' is removed?

> +		/* Avoid potential leak risk */

I think more detailed explanation about 'deadloop risk' and 'leak risk'
is necessary.

> +		case offsetof(struct smc_sock_negotiator_ops, release):
> +			return NULL;
> +		}
> +		return &bpf_sk_setsockopt_proto;
> +	case BPF_FUNC_getsockopt:
> +		return &bpf_sk_getsockopt_proto;
> +	case BPF_FUNC_skc_to_tcp_sock:
> +		return &bpf_smc_skc_to_tcp_sock_proto;
> +	default:
> +		return bpf_base_func_proto(func_id);
> +	}
> +}
> +
> +static bool smc_negotiator_prog_is_valid_access(int off, int size, enum bpf_access_type type,
> +						const struct bpf_prog *prog,
> +						struct bpf_insn_access_aux *info)
> +{
> +	if (!bpf_tracing_btf_ctx_access(off, size, type, prog, info))
> +		return false;
> +
> +	/* promote it to smc_sock */
> +	if (base_type(info->reg_type) == PTR_TO_BTF_ID &&
> +	    !bpf_type_has_unsafe_modifiers(info->reg_type) &&
> +	    info->btf_id == sock_id)
> +		info->btf_id = smc_sock_id;
> +
> +	return true;
> +}
> +
> +static const struct bpf_verifier_ops bpf_smc_negotiator_verifier_ops = {
> +	.get_func_proto  = smc_negotiator_prog_func_proto,
> +	.is_valid_access = smc_negotiator_prog_is_valid_access,
> +};
> +
> +struct bpf_struct_ops bpf_smc_sock_negotiator_ops = {
> +	.verifier_ops = &bpf_smc_negotiator_verifier_ops,
> +	.init = bpf_smc_negotiator_init,
> +	.check_member = bpf_smc_negotiator_check_member,
> +	.init_member = bpf_smc_negotiator_init_member,
> +	.reg = bpf_smc_negotiator_reg,
> +	.update = bpf_smc_negotiator_update,
> +	.unreg = bpf_smc_negotiator_unreg,
> +	.validate = bpf_smc_negotiator_validate,
> +	.name = "smc_sock_negotiator_ops",
> +};
> \ No newline at end of file

Empty line at the end?

Powered by blists - more mailing lists