netdev - RE: [Patch bpf v2 3/4] net: implement ->sock_is_readable for UDP and AF

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <61562fa0b3f74_6c4e4208c1@john-XPS-13-9370.notmuch>
Date:   Thu, 30 Sep 2021 14:44:00 -0700
From:   John Fastabend <john.fastabend@...il.com>
To:     Cong Wang <xiyou.wangcong@...il.com>, netdev@...r.kernel.org
Cc:     bpf@...r.kernel.org, Cong Wang <cong.wang@...edance.com>,
        Yucong Sun <sunyucong@...il.com>,
        John Fastabend <john.fastabend@...il.com>,
        Daniel Borkmann <daniel@...earbox.net>,
        Jakub Sitnicki <jakub@...udflare.com>,
        Lorenz Bauer <lmb@...udflare.com>
Subject: RE: [Patch bpf v2 3/4] net: implement ->sock_is_readable for UDP and
 AF_UNIX

Cong Wang wrote:
> From: Cong Wang <cong.wang@...edance.com>
> 
> Yucong noticed we can't poll() sockets in sockmap even when
> they are the destination sockets of redirections. This is
> because we never poll any psock queues in ->poll(), except
> for TCP. Now we can overwrite >sock_is_readable() and
> implement and invoke it for UDP and AF_UNIX sockets.

nit: instead of 'because we never poll any psock queue...' how about
'because we do not poll the psock queues in ->poll(), except
for TCP.'
> 
> Reported-by: Yucong Sun <sunyucong@...il.com>
> Cc: John Fastabend <john.fastabend@...il.com>
> Cc: Daniel Borkmann <daniel@...earbox.net>
> Cc: Jakub Sitnicki <jakub@...udflare.com>
> Cc: Lorenz Bauer <lmb@...udflare.com>
> Signed-off-by: Cong Wang <cong.wang@...edance.com>
> ---

[...]
  
>  static inline void sk_msg_check_to_free(struct sk_msg *msg, u32 i, u32 bytes)
>  {
> diff --git a/net/core/skmsg.c b/net/core/skmsg.c
> index 2d6249b28928..93ae48581ad2 100644
> --- a/net/core/skmsg.c
> +++ b/net/core/skmsg.c
> @@ -474,6 +474,20 @@ int sk_msg_recvmsg(struct sock *sk, struct sk_psock *psock, struct msghdr *msg,
>  }
>  EXPORT_SYMBOL_GPL(sk_msg_recvmsg);
>  
> +bool sk_msg_is_readable(struct sock *sk)
> +{
> +	struct sk_psock *psock;
> +	bool empty = true;
> +
> +	psock = sk_psock_get_checked(sk);

We shouldn't need the checked version here right? We only get here because
we hooked the sk with the callbacks from *_bpf_rebuild_rpotos. Then we
can just use sk_psock() and save a few extra insns/branch.

> +	if (IS_ERR_OR_NULL(psock))
> +		return false;
> +	empty = sk_psock_queue_empty(psock);
> +	sk_psock_put(sk, psock);
> +	return !empty;
> +}
> +EXPORT_SYMBOL_GPL(sk_msg_is_readable);

[...]