netdev - Re: [PATCH net] af_unix: Support POLLPRI for OOB.

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: <14f3c8e3-1b8f-7152-224b-6a4c9a0b6e61@gmail.com>
Date:   Mon, 14 Mar 2022 17:26:54 -0700
From:   Eric Dumazet <eric.dumazet@...il.com>
To:     Shoaib Rao <rao.shoaib@...cle.com>,
        Kuniyuki Iwashima <kuniyu@...zon.co.jp>,
        "David S. Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>
Cc:     Kuniyuki Iwashima <kuni1840@...il.com>, netdev@...r.kernel.org
Subject: Re: [PATCH net] af_unix: Support POLLPRI for OOB.


On 3/14/22 11:10, Shoaib Rao wrote:
>
> On 3/14/22 10:42, Eric Dumazet wrote:
>>
>> On 3/13/22 22:21, Kuniyuki Iwashima wrote:
>>> The commit 314001f0bf92 ("af_unix: Add OOB support") introduced OOB for
>>> AF_UNIX, but it lacks some changes for POLLPRI.  Let's add the missing
>>> piece.
>>>
>>> In the selftest, normal datagrams are sent followed by OOB data, so 
>>> this
>>> commit replaces `POLLIN | POLLPRI` with just `POLLPRI` in the first 
>>> test
>>> case.
>>>
>>> Fixes: 314001f0bf92 ("af_unix: Add OOB support")
>>> Signed-off-by: Kuniyuki Iwashima <kuniyu@...zon.co.jp>
>>> ---
>>>   net/unix/af_unix.c                                  | 2 ++
>>>   tools/testing/selftests/net/af_unix/test_unix_oob.c | 6 +++---
>>>   2 files changed, 5 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
>>> index c19569819866..711d21b1c3e1 100644
>>> --- a/net/unix/af_unix.c
>>> +++ b/net/unix/af_unix.c
>>> @@ -3139,6 +3139,8 @@ static __poll_t unix_poll(struct file *file, 
>>> struct socket *sock, poll_table *wa
>>>           mask |= EPOLLIN | EPOLLRDNORM;
>>>       if (sk_is_readable(sk))
>>>           mask |= EPOLLIN | EPOLLRDNORM;
>>> +    if (unix_sk(sk)->oob_skb)
>>> +        mask |= EPOLLPRI;
>>
>>
>> This adds another data-race, maybe add something to avoid another 
>> syzbot report ?
>
> It's not obvious to me how there would be a race as it is just a check.
>

KCSAN will detect that whenever unix_poll() reads oob_skb,

its value can be changed by another cpu.


unix_poll() runs without holding a lock.



> Also this change should be under #if IS_ENABLED(CONFIG_AF_UNIX_OOB)
>
> Thanks,
>
> Shoaib
>
>>
>>
>>>         /* Connection-based need to check for termination and 
>>> startup */
>>>       if ((sk->sk_type == SOCK_STREAM || sk->sk_type == 
>>> SOCK_SEQPACKET) &&
>>> diff --git a/tools/testing/selftests/net/af_unix/test_unix_oob.c 
>>> b/tools/testing/selftests/net/af_unix/test_unix_oob.c
>>> index 3dece8b29253..b57e91e1c3f2 100644
>>> --- a/tools/testing/selftests/net/af_unix/test_unix_oob.c
>>> +++ b/tools/testing/selftests/net/af_unix/test_unix_oob.c
>>> @@ -218,10 +218,10 @@ main(int argc, char **argv)
>>>         /* Test 1:
>>>        * veriyf that SIGURG is
>>> -     * delivered and 63 bytes are
>>> -     * read and oob is '@'
>>> +     * delivered, 63 bytes are
>>> +     * read, oob is '@', and POLLPRI works.
>>>        */
>>> -    wait_for_data(pfd, POLLIN | POLLPRI);
>>> +    wait_for_data(pfd, POLLPRI);
>>>       read_oob(pfd, &oob);
>>>       len = read_data(pfd, buf, 1024);
>>>       if (!signal_recvd || len != 63 || oob != '@') {