netdev - Re: [PATCH net,v2] tcp: make sure init the accept

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: <a0adea69-aea1-d35a-6a34-e1544a7ce1a5@huawei.com>
Date: Wed, 17 Jan 2024 10:48:53 +0800
From: shaozhengchao <shaozhengchao@...wei.com>
To: Eric Dumazet <edumazet@...gle.com>
CC: <netdev@...r.kernel.org>, <davem@...emloft.net>, <kuba@...nel.org>,
	<pabeni@...hat.com>, <dsahern@...nel.org>, <hkchu@...gle.com>,
	<weiyongjun1@...wei.com>, <yuehaibing@...wei.com>
Subject: Re: [PATCH net,v2] tcp: make sure init the accept_queue's spinlocks
 once



On 2024/1/16 18:38, Eric Dumazet wrote:
> On Sat, Jan 13, 2024 at 3:57 AM Zhengchao Shao <shaozhengchao@...wei.com> wrote:
>>
>> When I run syz's reproduction C program locally, it causes the following
>> issue:
>> pvqspinlock: lock 0xffff9d181cd5c660 has corrupted value 0x0!
>> WARNING: CPU: 19 PID: 21160 at __pv_queued_spin_unlock_slowpath (kernel/locking/qspinlock_paravirt.h:508)
>> Ha
> 
> 
>> When the socket receives the ACK packet during the three-way handshake,
>> it will hold spinlock. And then the user actively shutdowns the socket
>> and listens to the socket immediately, the spinlock will be initialized.
>> When the socket is going to release the spinlock, a warning is generated.
>> Also the same issue to fastopenq.lock.
>>
>> Add 'init_done' to make sure init the accept_queue's spinlocks once.
>>
>> Fixes: fff1f3001cc5 ("tcp: add a spinlock to protect struct request_sock_queue")
>> Fixes: 168a8f58059a ("tcp: TCP Fast Open Server - main code path")
>> Signed-off-by: Zhengchao Shao <shaozhengchao@...wei.com>
>> ---
>> v2: Add 'init_done' to make sure init the accept_queue's spinlocks once.
>> ---
>>   include/net/request_sock.h | 1 +
>>   net/core/request_sock.c    | 7 +++++--
>>   net/ipv4/tcp.c             | 1 +
>>   3 files changed, 7 insertions(+), 2 deletions(-)
>>
>> diff --git a/include/net/request_sock.h b/include/net/request_sock.h
>> index 144c39db9898..0054746fe92d 100644
>> --- a/include/net/request_sock.h
>> +++ b/include/net/request_sock.h
>> @@ -175,6 +175,7 @@ struct fastopen_queue {
>>   struct request_sock_queue {
>>          spinlock_t              rskq_lock;
>>          u8                      rskq_defer_accept;
>> +       bool                    init_done;
>>
> 
> No, we should not add a new field for this.
> The idea of having a conditional  spin_lock_init() is not very nice
> for code readability.
> 
> Just always init request_sock_queue spinlocks for all inet sockets at
> socket() and accept() time,
> not at listen() time.
> 
> This structure is not dynamically allocated, and part of 'struct
> inet_connection_sock'...
Hi Eric:
	It looks good to me. I will send V3.
Thank you.

Zhengchao Shao