Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Wed, 30 Nov 2022 05:18:20 -0800
From:   Breno Leitao <leitao@debian.org>
To:     "Iwashima, Kuniyuki" <kuniyu@amazon.co.jp>
Cc:     "davem@davemloft.net" <davem@davemloft.net>,
        "dsahern@kernel.org" <dsahern@kernel.org>,
        "edumazet@google.com" <edumazet@google.com>,
        "kuba@kernel.org" <kuba@kernel.org>, "leit@fb.com" <leit@fb.com>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "netdev@vger.kernel.org" <netdev@vger.kernel.org>,
        "pabeni@redhat.com" <pabeni@redhat.com>,
        "yoshfuji@linux-ipv6.org" <yoshfuji@linux-ipv6.org>
Subject: Re: [PATCH RESEND net-next] tcp: socket-specific version of
 WARN_ON_ONCE()
Message-ID: <Y4dYHLMTtr+QqeLm@gmail.com>
References: <20221124112229.789975-1-leitao@debian.org>
 <20221129010055.75780-1-kuniyu@amazon.com>
 <Y4X/XidkaLaD5Zak@gmail.com>
 <A5D7EBCF-CC49-4575-9DA7-0419BA1F0E9B@amazon.co.jp>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <A5D7EBCF-CC49-4575-9DA7-0419BA1F0E9B@amazon.co.jp>
Precedence: bulk

On Tue, Nov 29, 2022 at 09:16:16PM +0000, Iwashima, Kuniyuki wrote:
> > On Nov 29, 2022, at 21:48, Breno Leitao <leitao@debian.org> wrote:
> >> On Tue, Nov 29, 2022 at 10:00:55AM +0900, Kuniyuki Iwashima wrote:

<snip>

> >>> +void tcp_sock_warn(const struct tcp_sock *tp)
> >>> +{
> >>> +   const struct sock *sk = (const struct sock *)tp;
> >>> +   struct inet_sock *inet = inet_sk(sk);
> >>> +   struct inet_connection_sock *icsk = inet_csk(sk);
> >>> +
> >>> +   WARN_ON(1);
> >>> +
> >>> +   if (!tp)
> >> 
> >> Is this needed ?
> > 
> > We are de-referencing tp/sk in the lines below, so, I think it is safe to
> > check if they are not NULL before the de-refencing it.
> 
> tp->snd_cwnd is accessed just after this WARN, 
> so I thought there were no cases where tp is NULL.

Oh, important to say that we want to re-use this macro on other places
as well. This initial usage (on tcp_snd_cwnd_set()) is just for the
initial patch. I see value replacing some WARN_ON_*() by
TCP_SOCK_WARN_ON_ONCE() in other parts of the code, so, this check is to
protect this warning when TCP_SOCK_WARN_ON_ONCE() is called from
different places.

Anyway, I definitely can remove the check here, but, we might want to
re-add it later, as we replace some WARN_ON_* by TCP_SOCK_WARN_ON_*();

> I think this additional if could confuse future readers and 
> want to make sure if there is such a case.

How come checking if a pointer is valid before de-refencing it could
confuse readers?

Thank you for the review!