2021-03-29 22:25:49

by Sasha Levin

[permalink] [raw]
Subject: [PATCH AUTOSEL 5.10 09/33] net: correct sk_acceptq_is_full()

From: liuyacan <[email protected]>

[ Upstream commit f211ac154577ec9ccf07c15f18a6abf0d9bdb4ab ]

The "backlog" argument in listen() specifies
the maximom length of pending connections,
so the accept queue should be considered full
if there are exactly "backlog" elements.

Signed-off-by: liuyacan <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
Signed-off-by: Sasha Levin <[email protected]>
---
include/net/sock.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/net/sock.h b/include/net/sock.h
index 253202dcc5e6..8cdc009d7636 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -930,7 +930,7 @@ static inline void sk_acceptq_added(struct sock *sk)

static inline bool sk_acceptq_is_full(const struct sock *sk)
{
- return READ_ONCE(sk->sk_ack_backlog) > READ_ONCE(sk->sk_max_ack_backlog);
+ return READ_ONCE(sk->sk_ack_backlog) >= READ_ONCE(sk->sk_max_ack_backlog);
}

/*
--
2.30.1


2021-04-06 08:46:08

by Marcelo Ricardo Leitner

[permalink] [raw]
Subject: Re: [PATCH AUTOSEL 5.10 09/33] net: correct sk_acceptq_is_full()

On Mon, Mar 29, 2021 at 06:21:57PM -0400, Sasha Levin wrote:
> From: liuyacan <[email protected]>
>
> [ Upstream commit f211ac154577ec9ccf07c15f18a6abf0d9bdb4ab ]
>
> The "backlog" argument in listen() specifies
> the maximom length of pending connections,
> so the accept queue should be considered full
> if there are exactly "backlog" elements.

Hi Sasha. Can you please confirm that this one was dropped as well?
Thanks.

2021-04-07 21:04:08

by Sasha Levin

[permalink] [raw]
Subject: Re: [PATCH AUTOSEL 5.10 09/33] net: correct sk_acceptq_is_full()

On Mon, Apr 05, 2021 at 03:59:02PM -0300, Marcelo Ricardo Leitner wrote:
>On Mon, Mar 29, 2021 at 06:21:57PM -0400, Sasha Levin wrote:
>> From: liuyacan <[email protected]>
>>
>> [ Upstream commit f211ac154577ec9ccf07c15f18a6abf0d9bdb4ab ]
>>
>> The "backlog" argument in listen() specifies
>> the maximom length of pending connections,
>> so the accept queue should be considered full
>> if there are exactly "backlog" elements.
>
>Hi Sasha. Can you please confirm that this one was dropped as well?
>Thanks.

Yup!

--
Thanks,
Sasha