From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-7.0 required=3.0 tests=DKIMWL_WL_HIGH,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH, MAILING_LIST_MULTI,SIGNED_OFF_BY,SPF_HELO_NONE,SPF_PASS autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 2649CC54E8D for ; Tue, 12 May 2020 13:52:57 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id F392E20730 for ; Tue, 12 May 2020 13:52:56 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=cloudflare.com header.i=@cloudflare.com header.b="EscL7hFL" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1729637AbgELNw4 (ORCPT ); Tue, 12 May 2020 09:52:56 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:60706 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726891AbgELNw4 (ORCPT ); Tue, 12 May 2020 09:52:56 -0400 Received: from mail-wm1-x344.google.com (mail-wm1-x344.google.com [IPv6:2a00:1450:4864:20::344]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id E572DC061A0F for ; Tue, 12 May 2020 06:52:55 -0700 (PDT) Received: by mail-wm1-x344.google.com with SMTP id u16so23385160wmc.5 for ; Tue, 12 May 2020 06:52:55 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=cloudflare.com; s=google; h=references:user-agent:from:to:cc:subject:in-reply-to:date :message-id:mime-version; bh=Zq55omEuEvgB9vlSmywwcqU/9eUwhdaCINvvFlYpd7E=; b=EscL7hFLj3hSJ0yVelftgSOk+dLKEz8ujOtrGZt82du5ydK12x0HLwgQ1seiNhZlQj G4/hllre3xV+iTH2jbQYAzOKgTNhhrGFOL5C0MNfKXXmxcsNXEEbOyi87HI/Juh7wxV9 25+paHW711mNGDu/qwlz/0JhtlO0R9XCcjPHk= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:references:user-agent:from:to:cc:subject :in-reply-to:date:message-id:mime-version; bh=Zq55omEuEvgB9vlSmywwcqU/9eUwhdaCINvvFlYpd7E=; b=Rd7u9lGfeZwMNYjj3MsjPTApfGXIa7m4Wu0ASWRfZ2Iscde5LQ3YPq6a4kkSmgbMRK vwSzYbCzeQJAb8IRgyJ2EmDTDU8FYnA/dwnMTFQvYsagQUX8701n1lteb7kPfBym23jA GgMApQ+lUGqN6Mncik/oP7uIj+9Zm9HuxVsgeip+k03jJD+1DiEMA52Q7Eg/OyekdWj1 FiFfVHTtG5Xqr8aFgObd1GYrsz1TC1VKkI/JaWCzZQ1GDvpmkxCg4w/VNjFNAkQ5B0ui 0Qne2ckw09zX7LPvwzcmzxgwI3YAJljdqbTZAX9dsLIuTeqRlj611LFMs2dE9BW9HXEz EFEA== X-Gm-Message-State: AGi0PuZPXTtJPNJ/t9geGrtaY+ZpYo4q6a1hDtEtejROvEjU9UaS8o2+ mdgateRNNwsO7pJojcMh0GVDTw== X-Google-Smtp-Source: APiQypIhYlNCB706JpgHSgcF3mhOAfg+a2V0Z+Bq56eJiqRl08HhFazBmDZM8CGGj0okCR8+qdPEUA== X-Received: by 2002:a1c:2702:: with SMTP id n2mr12984805wmn.107.1589291574318; Tue, 12 May 2020 06:52:54 -0700 (PDT) Received: from cloudflare.com ([2a02:a310:c262:aa00:b35e:8938:2c2a:ba8b]) by smtp.gmail.com with ESMTPSA id s2sm326571wme.33.2020.05.12.06.52.53 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 12 May 2020 06:52:53 -0700 (PDT) References: <20200511185218.1422406-1-jakub@cloudflare.com> <20200511185218.1422406-6-jakub@cloudflare.com> <20200511204445.i7sessmtszox36xd@ast-mbp> User-agent: mu4e 1.1.0; emacs 26.3 From: Jakub Sitnicki To: Alexei Starovoitov Cc: netdev@vger.kernel.org, bpf@vger.kernel.org, dccp@vger.kernel.org, kernel-team@cloudflare.com, Alexei Starovoitov , Daniel Borkmann , "David S. Miller" , Eric Dumazet , Gerrit Renker , Jakub Kicinski , Andrii Nakryiko , Martin KaFai Lau , Marek Majkowski , Lorenz Bauer Subject: Re: [PATCH bpf-next v2 05/17] inet: Run SK_LOOKUP BPF program on socket lookup In-reply-to: <20200511204445.i7sessmtszox36xd@ast-mbp> Date: Tue, 12 May 2020 15:52:52 +0200 Message-ID: <871rnpuuob.fsf@cloudflare.com> MIME-Version: 1.0 Content-Type: text/plain Sender: bpf-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: bpf@vger.kernel.org On Mon, May 11, 2020 at 10:44 PM CEST, Alexei Starovoitov wrote: > On Mon, May 11, 2020 at 08:52:06PM +0200, Jakub Sitnicki wrote: >> Run a BPF program before looking up a listening socket on the receive path. >> Program selects a listening socket to yield as result of socket lookup by >> calling bpf_sk_assign() helper and returning BPF_REDIRECT code. >> >> Alternatively, program can also fail the lookup by returning with BPF_DROP, >> or let the lookup continue as usual with BPF_OK on return. >> >> This lets the user match packets with listening sockets freely at the last >> possible point on the receive path, where we know that packets are destined >> for local delivery after undergoing policing, filtering, and routing. >> >> With BPF code selecting the socket, directing packets destined to an IP >> range or to a port range to a single socket becomes possible. >> >> Suggested-by: Marek Majkowski >> Reviewed-by: Lorenz Bauer >> Signed-off-by: Jakub Sitnicki >> --- >> include/net/inet_hashtables.h | 36 +++++++++++++++++++++++++++++++++++ >> net/ipv4/inet_hashtables.c | 15 ++++++++++++++- >> 2 files changed, 50 insertions(+), 1 deletion(-) >> >> diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h >> index 6072dfbd1078..3fcbc8f66f88 100644 >> --- a/include/net/inet_hashtables.h >> +++ b/include/net/inet_hashtables.h >> @@ -422,4 +422,40 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, >> >> int inet_hash_connect(struct inet_timewait_death_row *death_row, >> struct sock *sk); >> + >> +static inline struct sock *bpf_sk_lookup_run(struct net *net, >> + struct bpf_sk_lookup_kern *ctx) >> +{ >> + struct bpf_prog *prog; >> + int ret = BPF_OK; >> + >> + rcu_read_lock(); >> + prog = rcu_dereference(net->sk_lookup_prog); >> + if (prog) >> + ret = BPF_PROG_RUN(prog, ctx); >> + rcu_read_unlock(); >> + >> + if (ret == BPF_DROP) >> + return ERR_PTR(-ECONNREFUSED); >> + if (ret == BPF_REDIRECT) >> + return ctx->selected_sk; >> + return NULL; >> +} >> + >> +static inline struct sock *inet_lookup_run_bpf(struct net *net, u8 protocol, >> + __be32 saddr, __be16 sport, >> + __be32 daddr, u16 dport) >> +{ >> + struct bpf_sk_lookup_kern ctx = { >> + .family = AF_INET, >> + .protocol = protocol, >> + .v4.saddr = saddr, >> + .v4.daddr = daddr, >> + .sport = sport, >> + .dport = dport, >> + }; >> + >> + return bpf_sk_lookup_run(net, &ctx); >> +} >> + >> #endif /* _INET_HASHTABLES_H */ >> diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c >> index ab64834837c8..f4d07285591a 100644 >> --- a/net/ipv4/inet_hashtables.c >> +++ b/net/ipv4/inet_hashtables.c >> @@ -307,9 +307,22 @@ struct sock *__inet_lookup_listener(struct net *net, >> const int dif, const int sdif) >> { >> struct inet_listen_hashbucket *ilb2; >> - struct sock *result = NULL; >> + struct sock *result, *reuse_sk; >> unsigned int hash2; >> >> + /* Lookup redirect from BPF */ >> + result = inet_lookup_run_bpf(net, hashinfo->protocol, >> + saddr, sport, daddr, hnum); >> + if (IS_ERR(result)) >> + return NULL; >> + if (result) { >> + reuse_sk = lookup_reuseport(net, result, skb, doff, >> + saddr, sport, daddr, hnum); >> + if (reuse_sk) >> + result = reuse_sk; >> + goto done; >> + } >> + > > The overhead is too high to do this all the time. > The feature has to be static_key-ed. Static keys is something that Lorenz has also suggested internally, but we wanted to keep it simple at first. Introduction of static keys forces us to decide when non-init_net netns are allowed to attach to SK_LOOKUP, as attaching enabling SK_LOOKUP in isolated netns will affect the rx path in init_net. I see two options, which seem sensible: 1) limit SK_LOOKUP to init_net, which makes testing setup harder, or 2) allow non-init_net netns to attach to SK_LOOKUP only if static key has been already enabled (via sysctl?). > > Also please add multi-prog support. Adding it later will cause > all sorts of compatibility issues. The semantics of multi-prog > needs to be thought through right now. > For example BPF_DROP or BPF_REDIRECT could terminate the prog_run_array > sequence of progs while BPF_OK could continue. > It's not ideal, but better than nothing. I must say this approach is quite appealing because it's simple to explain. I would need a custom BPF_PROG_RUN_ARRAY, though. I'm curious what downside do you see here? Is overriding an earlier DROP/REDIRECT verdict useful? > Another option could be to execute all attached progs regardless > of return code, but don't let second prog override selected_sk blindly. > bpf_sk_assign() could get smarter. So if IIUC the rough idea here would be like below? - 1st program calls bpf_sk_assign(ctx, sk1, 0 /*flags*/) -> 0 (OK) - 2nd program calls bpf_sk_assign(ctx, sk2, 0) -> -EBUSY (already selected) bpf_sk_assign(ctx, sk2, BPF_EXIST) -> 0 (OK, replace existing) In this case the last program to run has the final say, as opposed to the semantics where DROP/REDIRECT terminates. Also, 2nd and subsequent programs would probably need to know if and which socket has been already selected. I think the selection could be exposed in context as bpf_sock pointer. I admit, I can't quite see the benefit of running thru all programs in array, so I'm tempted to go with terminate of DROP/REDIRECT in v3. > > Also please switch to bpf_link way of attaching. All system wide attachments > should be visible and easily debuggable via 'bpftool link show'. > Currently we're converting tc and xdp hooks to bpf_link. This new hook > should have it from the beginning. Will do in v3. Thanks for feedback, Jakub From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jakub Sitnicki Date: Tue, 12 May 2020 13:52:52 +0000 Subject: Re: [PATCH bpf-next v2 05/17] inet: Run SK_LOOKUP BPF program on socket lookup Message-Id: <871rnpuuob.fsf@cloudflare.com> List-Id: References: <20200511185218.1422406-6-jakub@cloudflare.com> In-Reply-To: <20200511185218.1422406-6-jakub@cloudflare.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: dccp@vger.kernel.org On Mon, May 11, 2020 at 10:44 PM CEST, Alexei Starovoitov wrote: > On Mon, May 11, 2020 at 08:52:06PM +0200, Jakub Sitnicki wrote: >> Run a BPF program before looking up a listening socket on the receive path. >> Program selects a listening socket to yield as result of socket lookup by >> calling bpf_sk_assign() helper and returning BPF_REDIRECT code. >> >> Alternatively, program can also fail the lookup by returning with BPF_DROP, >> or let the lookup continue as usual with BPF_OK on return. >> >> This lets the user match packets with listening sockets freely at the last >> possible point on the receive path, where we know that packets are destined >> for local delivery after undergoing policing, filtering, and routing. >> >> With BPF code selecting the socket, directing packets destined to an IP >> range or to a port range to a single socket becomes possible. >> >> Suggested-by: Marek Majkowski >> Reviewed-by: Lorenz Bauer >> Signed-off-by: Jakub Sitnicki >> --- >> include/net/inet_hashtables.h | 36 +++++++++++++++++++++++++++++++++++ >> net/ipv4/inet_hashtables.c | 15 ++++++++++++++- >> 2 files changed, 50 insertions(+), 1 deletion(-) >> >> diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h >> index 6072dfbd1078..3fcbc8f66f88 100644 >> --- a/include/net/inet_hashtables.h >> +++ b/include/net/inet_hashtables.h >> @@ -422,4 +422,40 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, >> >> int inet_hash_connect(struct inet_timewait_death_row *death_row, >> struct sock *sk); >> + >> +static inline struct sock *bpf_sk_lookup_run(struct net *net, >> + struct bpf_sk_lookup_kern *ctx) >> +{ >> + struct bpf_prog *prog; >> + int ret = BPF_OK; >> + >> + rcu_read_lock(); >> + prog = rcu_dereference(net->sk_lookup_prog); >> + if (prog) >> + ret = BPF_PROG_RUN(prog, ctx); >> + rcu_read_unlock(); >> + >> + if (ret = BPF_DROP) >> + return ERR_PTR(-ECONNREFUSED); >> + if (ret = BPF_REDIRECT) >> + return ctx->selected_sk; >> + return NULL; >> +} >> + >> +static inline struct sock *inet_lookup_run_bpf(struct net *net, u8 protocol, >> + __be32 saddr, __be16 sport, >> + __be32 daddr, u16 dport) >> +{ >> + struct bpf_sk_lookup_kern ctx = { >> + .family = AF_INET, >> + .protocol = protocol, >> + .v4.saddr = saddr, >> + .v4.daddr = daddr, >> + .sport = sport, >> + .dport = dport, >> + }; >> + >> + return bpf_sk_lookup_run(net, &ctx); >> +} >> + >> #endif /* _INET_HASHTABLES_H */ >> diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c >> index ab64834837c8..f4d07285591a 100644 >> --- a/net/ipv4/inet_hashtables.c >> +++ b/net/ipv4/inet_hashtables.c >> @@ -307,9 +307,22 @@ struct sock *__inet_lookup_listener(struct net *net, >> const int dif, const int sdif) >> { >> struct inet_listen_hashbucket *ilb2; >> - struct sock *result = NULL; >> + struct sock *result, *reuse_sk; >> unsigned int hash2; >> >> + /* Lookup redirect from BPF */ >> + result = inet_lookup_run_bpf(net, hashinfo->protocol, >> + saddr, sport, daddr, hnum); >> + if (IS_ERR(result)) >> + return NULL; >> + if (result) { >> + reuse_sk = lookup_reuseport(net, result, skb, doff, >> + saddr, sport, daddr, hnum); >> + if (reuse_sk) >> + result = reuse_sk; >> + goto done; >> + } >> + > > The overhead is too high to do this all the time. > The feature has to be static_key-ed. Static keys is something that Lorenz has also suggested internally, but we wanted to keep it simple at first. Introduction of static keys forces us to decide when non-init_net netns are allowed to attach to SK_LOOKUP, as attaching enabling SK_LOOKUP in isolated netns will affect the rx path in init_net. I see two options, which seem sensible: 1) limit SK_LOOKUP to init_net, which makes testing setup harder, or 2) allow non-init_net netns to attach to SK_LOOKUP only if static key has been already enabled (via sysctl?). > > Also please add multi-prog support. Adding it later will cause > all sorts of compatibility issues. The semantics of multi-prog > needs to be thought through right now. > For example BPF_DROP or BPF_REDIRECT could terminate the prog_run_array > sequence of progs while BPF_OK could continue. > It's not ideal, but better than nothing. I must say this approach is quite appealing because it's simple to explain. I would need a custom BPF_PROG_RUN_ARRAY, though. I'm curious what downside do you see here? Is overriding an earlier DROP/REDIRECT verdict useful? > Another option could be to execute all attached progs regardless > of return code, but don't let second prog override selected_sk blindly. > bpf_sk_assign() could get smarter. So if IIUC the rough idea here would be like below? - 1st program calls bpf_sk_assign(ctx, sk1, 0 /*flags*/) -> 0 (OK) - 2nd program calls bpf_sk_assign(ctx, sk2, 0) -> -EBUSY (already selected) bpf_sk_assign(ctx, sk2, BPF_EXIST) -> 0 (OK, replace existing) In this case the last program to run has the final say, as opposed to the semantics where DROP/REDIRECT terminates. Also, 2nd and subsequent programs would probably need to know if and which socket has been already selected. I think the selection could be exposed in context as bpf_sock pointer. I admit, I can't quite see the benefit of running thru all programs in array, so I'm tempted to go with terminate of DROP/REDIRECT in v3. > > Also please switch to bpf_link way of attaching. All system wide attachments > should be visible and easily debuggable via 'bpftool link show'. > Currently we're converting tc and xdp hooks to bpf_link. This new hook > should have it from the beginning. Will do in v3. Thanks for feedback, Jakub