[PATCH] bpf: Fix bpf_get/setsockopt to tos not take effect when TCP over IPv4 via INET6 API

[PATCH] bpf: Fix bpf_get/setsockopt to tos not take effect when TCP over IPv4 via INET6 API

[Feng zhou]

From: Feng Zhou <zhoufeng.zf@bytedance.com>

When TCP over IPv4 via INET6 API, bpf_get/setsockopt with ipv4 will
fail, because sk->sk_family is AF_INET6. With ipv6 will success, not
take effect, because inet_csk(sk)->icsk_af_ops is ipv6_mapped and
use ip_queue_xmit, inet_sk(sk)->tos.

So bpf_get/setsockopt needs add the judgment of this case. Just check
"inet_csk(sk)->icsk_af_ops == &ipv6_mapped".

Signed-off-by: Feng Zhou <zhoufeng.zf@bytedance.com>
---
 include/net/tcp.h   | 2 ++
 net/core/filter.c   | 2 +-
 net/ipv6/tcp_ipv6.c | 5 +++++
 3 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 2aac11e7e1cc..ea673f88c900 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -493,6 +493,8 @@ struct request_sock *cookie_tcp_reqsk_alloc(const struct request_sock_ops *ops,
 					    struct tcp_options_received *tcp_opt,
 					    int mss, u32 tsoff);
 
+bool is_tcp_sock_ipv6_mapped(struct sock *sk);
+
 #if IS_ENABLED(CONFIG_BPF)
 struct bpf_tcp_req_attrs {
 	u32 rcv_tsval;
diff --git a/net/core/filter.c b/net/core/filter.c
index 78a6f746ea0b..9798537044be 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -5399,7 +5399,7 @@ static int sol_ip_sockopt(struct sock *sk, int optname,
 			  char *optval, int *optlen,
 			  bool getopt)
 {
-	if (sk->sk_family != AF_INET)
+	if (sk->sk_family != AF_INET && !is_tcp_sock_ipv6_mapped(sk))
 		return -EINVAL;
 
 	switch (optname) {
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 200fea92f12f..84651d630c89 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -92,6 +92,11 @@ static const struct tcp_sock_af_ops tcp_sock_ipv6_mapped_specific;
 #define tcp_inet6_sk(sk) (&container_of_const(tcp_sk(sk), \
 					      struct tcp6_sock, tcp)->inet6)
 
+bool is_tcp_sock_ipv6_mapped(struct sock *sk)
+{
+	return (inet_csk(sk)->icsk_af_ops == &ipv6_mapped);
+}
+
 static void inet6_sk_rx_dst_set(struct sock *sk, const struct sk_buff *skb)
 {
 	struct dst_entry *dst = skb_dst(skb);
-- 
2.30.2

Re: [PATCH] bpf: Fix bpf_get/setsockopt to tos not take effect when TCP over IPv4 via INET6 API

[Sabrina Dubroca]

2024-08-14, 16:45:04 +0800, Feng zhou wrote:
> diff --git a/net/core/filter.c b/net/core/filter.c
> index 78a6f746ea0b..9798537044be 100644
> --- a/net/core/filter.c
> +++ b/net/core/filter.c
> @@ -5399,7 +5399,7 @@ static int sol_ip_sockopt(struct sock *sk, int optname,
>  			  char *optval, int *optlen,
>  			  bool getopt)
>  {
> -	if (sk->sk_family != AF_INET)
> +	if (sk->sk_family != AF_INET && !is_tcp_sock_ipv6_mapped(sk))
>  		return -EINVAL;

I don't think this works when CONFIG_IPV6=m, because then
is_tcp_sock_ipv6_mapped will be part of the module and not usable in
net/core/filter.c. Stuff like this is usually done through ipv6_stub.

-- 
Sabrina

Re: [PATCH] bpf: Fix bpf_get/setsockopt to tos not take effect when TCP over IPv4 via INET6 API

[kernel test robot]

Hi Feng,

kernel test robot noticed the following build errors:

[auto build test ERROR on bpf-next/master]
[also build test ERROR on bpf/master net-next/main net/main linus/master v6.11-rc3 next-20240815]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Feng-zhou/bpf-Fix-bpf_get-setsockopt-to-tos-not-take-effect-when-TCP-over-IPv4-via-INET6-API/20240814-231142
base:   https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git master
patch link:    https://lore.kernel.org/r/20240814084504.22172-1-zhoufeng.zf%40bytedance.com
patch subject: [PATCH] bpf: Fix bpf_get/setsockopt to tos not take effect when TCP over IPv4 via INET6 API
config: alpha-defconfig (https://download.01.org/0day-ci/archive/20240815/202408152034.lw9Ilsj6-lkp@intel.com/config)
compiler: alpha-linux-gcc (GCC) 13.3.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240815/202408152034.lw9Ilsj6-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202408152034.lw9Ilsj6-lkp@intel.com/

All errors (new ones prefixed by >>):

   alpha-linux-ld: net/core/filter.o: in function `sol_ip_sockopt':
>> net/core/filter.c:5402:(.text+0x8acc): undefined reference to `is_tcp_sock_ipv6_mapped'
>> alpha-linux-ld: net/core/filter.c:5402:(.text+0x8ad4): undefined reference to `is_tcp_sock_ipv6_mapped'
   alpha-linux-ld: net/core/filter.c:5402:(.text+0xc84c): undefined reference to `is_tcp_sock_ipv6_mapped'
   alpha-linux-ld: net/core/filter.c:5402:(.text+0xc858): undefined reference to `is_tcp_sock_ipv6_mapped'


vim +5402 net/core/filter.c

  5397	
  5398	static int sol_ip_sockopt(struct sock *sk, int optname,
  5399				  char *optval, int *optlen,
  5400				  bool getopt)
  5401	{
> 5402		if (sk->sk_family != AF_INET && !is_tcp_sock_ipv6_mapped(sk))
  5403			return -EINVAL;
  5404	
  5405		switch (optname) {
  5406		case IP_TOS:
  5407			if (*optlen != sizeof(int))
  5408				return -EINVAL;
  5409			break;
  5410		default:
  5411			return -EINVAL;
  5412		}
  5413	
  5414		if (getopt)
  5415			return do_ip_getsockopt(sk, SOL_IP, optname,
  5416						KERNEL_SOCKPTR(optval),
  5417						KERNEL_SOCKPTR(optlen));
  5418	
  5419		return do_ip_setsockopt(sk, SOL_IP, optname,
  5420					KERNEL_SOCKPTR(optval), *optlen);
  5421	}
  5422	

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki

Re: [PATCH] bpf: Fix bpf_get/setsockopt to tos not take effect when TCP over IPv4 via INET6 API

[kernel test robot]

Hi Feng,

kernel test robot noticed the following build errors:

[auto build test ERROR on bpf-next/master]
[also build test ERROR on bpf/master net-next/main net/main linus/master v6.11-rc3 next-20240815]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Feng-zhou/bpf-Fix-bpf_get-setsockopt-to-tos-not-take-effect-when-TCP-over-IPv4-via-INET6-API/20240814-231142
base:   https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git master
patch link:    https://lore.kernel.org/r/20240814084504.22172-1-zhoufeng.zf%40bytedance.com
patch subject: [PATCH] bpf: Fix bpf_get/setsockopt to tos not take effect when TCP over IPv4 via INET6 API
config: openrisc-defconfig (https://download.01.org/0day-ci/archive/20240815/202408152058.YXAnhLgZ-lkp@intel.com/config)
compiler: or1k-linux-gcc (GCC) 14.1.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240815/202408152058.YXAnhLgZ-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202408152058.YXAnhLgZ-lkp@intel.com/

All errors (new ones prefixed by >>):

   or1k-linux-ld: net/core/filter.o: in function `__bpf_getsockopt':
>> filter.c:(.text+0xa69c): undefined reference to `is_tcp_sock_ipv6_mapped'
>> filter.c:(.text+0xa69c): relocation truncated to fit: R_OR1K_INSN_REL_26 against undefined symbol `is_tcp_sock_ipv6_mapped'
   or1k-linux-ld: net/core/filter.o: in function `__bpf_setsockopt':
   filter.c:(.text+0xd660): undefined reference to `is_tcp_sock_ipv6_mapped'
   filter.c:(.text+0xd660): relocation truncated to fit: R_OR1K_INSN_REL_26 against undefined symbol `is_tcp_sock_ipv6_mapped'

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki