BPF List
 help / color / mirror / Atom feed
From: Martin KaFai Lau <kafai@fb.com>
To: <bpf@vger.kernel.org>, <netdev@vger.kernel.org>
Cc: Alexei Starovoitov <ast@kernel.org>,
	Andrii Nakryiko <andrii@kernel.org>,
	Daniel Borkmann <daniel@iogearbox.net>,
	David Miller <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	Jakub Kicinski <kuba@kernel.org>, <kernel-team@fb.com>,
	Paolo Abeni <pabeni@redhat.com>,
	Stanislav Fomichev <sdf@google.com>
Subject: [PATCH v2 bpf-next 03/15] bpf: net: Consider in_bpf() when testing capable() in sk_setsockopt()
Date: Wed, 3 Aug 2022 13:46:20 -0700	[thread overview]
Message-ID: <20220803204620.3077641-1-kafai@fb.com> (raw)
In-Reply-To: <20220803204601.3075863-1-kafai@fb.com>

When bpf program calling bpf_setsockopt(SOL_SOCKET),
it could be run in softirq and doesn't make sense to do the capable
check.  There was a similar situation in bpf_setsockopt(TCP_CONGESTION).
In commit 8d650cdedaab ("tcp: fix tcp_set_congestion_control() use from bpf hook"),
tcp_set_congestion_control(..., cap_net_admin) was added to skip
the cap check for bpf prog.

This patch adds sockopt_ns_capable() and sockopt_capable() for
the sk_setsockopt() to use.  They will consider the in_bpf()
before doing the ns_capable() and capable() test.  They are in
EXPORT_SYMBOL for the ipv6 module to use in a latter patch.

Suggested-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
---
 include/net/sock.h |  2 ++
 net/core/sock.c    | 38 +++++++++++++++++++++++++-------------
 2 files changed, 27 insertions(+), 13 deletions(-)

diff --git a/include/net/sock.h b/include/net/sock.h
index b2ff230860c6..72b78c2b6f83 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1723,6 +1723,8 @@ static inline void unlock_sock_fast(struct sock *sk, bool slow)
 
 void sockopt_lock_sock(struct sock *sk);
 void sockopt_release_sock(struct sock *sk);
+bool sockopt_ns_capable(struct user_namespace *ns, int cap);
+bool sockopt_capable(int cap);
 
 /* Used by processes to "lock" a socket state, so that
  * interrupts and bottom half handlers won't change it
diff --git a/net/core/sock.c b/net/core/sock.c
index 82759540ae2c..2d88c06c27b7 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1056,6 +1056,18 @@ void sockopt_release_sock(struct sock *sk)
 }
 EXPORT_SYMBOL(sockopt_release_sock);
 
+bool sockopt_ns_capable(struct user_namespace *ns, int cap)
+{
+	return in_bpf() || ns_capable(ns, cap);
+}
+EXPORT_SYMBOL(sockopt_ns_capable);
+
+bool sockopt_capable(int cap)
+{
+	return in_bpf() || capable(cap);
+}
+EXPORT_SYMBOL(sockopt_capable);
+
 /*
  *	This is meant for all protocols to use and covers goings on
  *	at the socket level. Everything here is generic.
@@ -1091,7 +1103,7 @@ static int sk_setsockopt(struct sock *sk, int level, int optname,
 
 	switch (optname) {
 	case SO_DEBUG:
-		if (val && !capable(CAP_NET_ADMIN))
+		if (val && !sockopt_capable(CAP_NET_ADMIN))
 			ret = -EACCES;
 		else
 			sock_valbool_flag(sk, SOCK_DBG, valbool);
@@ -1135,7 +1147,7 @@ static int sk_setsockopt(struct sock *sk, int level, int optname,
 		break;
 
 	case SO_SNDBUFFORCE:
-		if (!capable(CAP_NET_ADMIN)) {
+		if (!sockopt_capable(CAP_NET_ADMIN)) {
 			ret = -EPERM;
 			break;
 		}
@@ -1157,7 +1169,7 @@ static int sk_setsockopt(struct sock *sk, int level, int optname,
 		break;
 
 	case SO_RCVBUFFORCE:
-		if (!capable(CAP_NET_ADMIN)) {
+		if (!sockopt_capable(CAP_NET_ADMIN)) {
 			ret = -EPERM;
 			break;
 		}
@@ -1184,8 +1196,8 @@ static int sk_setsockopt(struct sock *sk, int level, int optname,
 
 	case SO_PRIORITY:
 		if ((val >= 0 && val <= 6) ||
-		    ns_capable(sock_net(sk)->user_ns, CAP_NET_RAW) ||
-		    ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN))
+		    sockopt_ns_capable(sock_net(sk)->user_ns, CAP_NET_RAW) ||
+		    sockopt_ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN))
 			sk->sk_priority = val;
 		else
 			ret = -EPERM;
@@ -1330,8 +1342,8 @@ static int sk_setsockopt(struct sock *sk, int level, int optname,
 			clear_bit(SOCK_PASSSEC, &sock->flags);
 		break;
 	case SO_MARK:
-		if (!ns_capable(sock_net(sk)->user_ns, CAP_NET_RAW) &&
-		    !ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) {
+		if (!sockopt_ns_capable(sock_net(sk)->user_ns, CAP_NET_RAW) &&
+		    !sockopt_ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) {
 			ret = -EPERM;
 			break;
 		}
@@ -1339,8 +1351,8 @@ static int sk_setsockopt(struct sock *sk, int level, int optname,
 		__sock_set_mark(sk, val);
 		break;
 	case SO_RCVMARK:
-		if (!ns_capable(sock_net(sk)->user_ns, CAP_NET_RAW) &&
-		    !ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) {
+		if (!sockopt_ns_capable(sock_net(sk)->user_ns, CAP_NET_RAW) &&
+		    !sockopt_ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) {
 			ret = -EPERM;
 			break;
 		}
@@ -1374,7 +1386,7 @@ static int sk_setsockopt(struct sock *sk, int level, int optname,
 #ifdef CONFIG_NET_RX_BUSY_POLL
 	case SO_BUSY_POLL:
 		/* allow unprivileged users to decrease the value */
-		if ((val > sk->sk_ll_usec) && !capable(CAP_NET_ADMIN))
+		if ((val > sk->sk_ll_usec) && !sockopt_capable(CAP_NET_ADMIN))
 			ret = -EPERM;
 		else {
 			if (val < 0)
@@ -1384,13 +1396,13 @@ static int sk_setsockopt(struct sock *sk, int level, int optname,
 		}
 		break;
 	case SO_PREFER_BUSY_POLL:
-		if (valbool && !capable(CAP_NET_ADMIN))
+		if (valbool && !sockopt_capable(CAP_NET_ADMIN))
 			ret = -EPERM;
 		else
 			WRITE_ONCE(sk->sk_prefer_busy_poll, valbool);
 		break;
 	case SO_BUSY_POLL_BUDGET:
-		if (val > READ_ONCE(sk->sk_busy_poll_budget) && !capable(CAP_NET_ADMIN)) {
+		if (val > READ_ONCE(sk->sk_busy_poll_budget) && !sockopt_capable(CAP_NET_ADMIN)) {
 			ret = -EPERM;
 		} else {
 			if (val < 0 || val > U16_MAX)
@@ -1461,7 +1473,7 @@ static int sk_setsockopt(struct sock *sk, int level, int optname,
 		 * scheduler has enough safe guards.
 		 */
 		if (sk_txtime.clockid != CLOCK_MONOTONIC &&
-		    !ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) {
+		    !sockopt_ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) {
 			ret = -EPERM;
 			break;
 		}
-- 
2.30.2


  parent reply	other threads:[~2022-08-03 20:49 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-08-03 20:46 [PATCH v2 bpf-next 00/15] bpf: net: Remove duplicated code from bpf_setsockopt() Martin KaFai Lau
2022-08-03 20:46 ` [PATCH v2 bpf-next 01/15] net: Add sk_setsockopt() to take the sk ptr instead of the sock ptr Martin KaFai Lau
2022-08-03 20:46 ` [PATCH v2 bpf-next 02/15] bpf: net: Avoid sk_setsockopt() taking sk lock when called from bpf Martin KaFai Lau
2022-08-03 22:59   ` sdf
2022-08-03 23:19     ` Martin KaFai Lau
2022-08-03 23:24       ` Stanislav Fomichev
2022-08-03 23:35         ` Martin KaFai Lau
2022-08-04 19:03   ` Andrii Nakryiko
2022-08-04 19:29     ` Martin KaFai Lau
2022-08-04 20:51       ` Universally available bpf_ctx WAS: " Andrii Nakryiko
2022-08-04 21:43         ` Stanislav Fomichev
2022-08-05  0:29         ` Martin KaFai Lau
2022-08-03 20:46 ` Martin KaFai Lau [this message]
2022-08-03 20:46 ` [PATCH v2 bpf-next 04/15] bpf: net: Change do_tcp_setsockopt() to use the sockopt's lock_sock() and capable() Martin KaFai Lau
2022-08-03 20:46 ` [PATCH v2 bpf-next 05/15] bpf: net: Change do_ip_setsockopt() " Martin KaFai Lau
2022-08-03 20:46 ` [PATCH v2 bpf-next 06/15] bpf: net: Change do_ipv6_setsockopt() " Martin KaFai Lau
2022-08-03 20:46 ` [PATCH v2 bpf-next 07/15] bpf: Initialize the bpf_run_ctx in bpf_iter_run_prog() Martin KaFai Lau
2022-08-03 20:46 ` [PATCH v2 bpf-next 08/15] bpf: Embed kernel CONFIG check into the if statement in bpf_setsockopt Martin KaFai Lau
2022-08-03 20:46 ` [PATCH v2 bpf-next 09/15] bpf: Change bpf_setsockopt(SOL_SOCKET) to reuse sk_setsockopt() Martin KaFai Lau
2022-08-03 20:47 ` [PATCH v2 bpf-next 10/15] bpf: Refactor bpf specific tcp optnames to a new function Martin KaFai Lau
2022-08-03 20:47 ` [PATCH v2 bpf-next 11/15] bpf: Change bpf_setsockopt(SOL_TCP) to reuse do_tcp_setsockopt() Martin KaFai Lau
2022-08-03 20:47 ` [PATCH v2 bpf-next 12/15] bpf: Change bpf_setsockopt(SOL_IP) to reuse do_ip_setsockopt() Martin KaFai Lau
2022-08-03 20:47 ` [PATCH v2 bpf-next 13/15] bpf: Change bpf_setsockopt(SOL_IPV6) to reuse do_ipv6_setsockopt() Martin KaFai Lau
2022-08-03 20:47 ` [PATCH v2 bpf-next 14/15] bpf: Add a few optnames to bpf_setsockopt Martin KaFai Lau
2022-08-03 20:47 ` [PATCH v2 bpf-next 15/15] selftests/bpf: bpf_setsockopt tests Martin KaFai Lau
2022-08-03 23:30   ` sdf
2022-08-04  0:04     ` Martin KaFai Lau
2022-08-04 17:03       ` Stanislav Fomichev
2022-08-04 19:17         ` Martin KaFai Lau

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220803204620.3077641-1-kafai@fb.com \
    --to=kafai@fb.com \
    --cc=andrii@kernel.org \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=kernel-team@fb.com \
    --cc=kuba@kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=sdf@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox