[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200411182813.GA18221@duo.ucw.cz>
Date: Sat, 11 Apr 2020 20:28:13 +0200
From: Pavel Machek <pavel@...x.de>
To: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc: linux-kernel@...r.kernel.org, stable@...r.kernel.org,
Qiujun Huang <hqjagain@...il.com>,
Marcelo Ricardo Leitner <mleitner@...hat.com>,
"David S. Miller" <davem@...emloft.net>,
syzbot+cea71eec5d6de256d54d@...kaller.appspotmail.com
Subject: Re: [PATCH 4.19 03/54] sctp: fix refcount bug in sctp_wfree
Hi!
> The following case cause the bug:
> for the trouble SKB, it was in outq->transmitted list
Ok... but this is one hell of "interesting" code.
> --- a/net/sctp/socket.c
> +++ b/net/sctp/socket.c
> @@ -165,29 +165,44 @@ static void sctp_clear_owner_w(struct sc
> skb_orphan(chunk->skb);
> }
>
> +#define traverse_and_process() \
> +do { \
> + msg = chunk->msg; \
> + if (msg == prev_msg) \
> + continue; \
> + list_for_each_entry(c, &msg->chunks, frag_list) { \
> + if ((clear && asoc->base.sk == c->skb->sk) || \
> + (!clear && asoc->base.sk != c->skb->sk)) \
> + cb(c); \
> + } \
> + prev_msg = msg; \
> +} while (0)
Should this be function?
Do you see how it does "continue"? But the do {} while(0) wrapper eats
the continue. "break" would be more clear here, but they are really
equivalent due to way this macro is used.
It is just very, very confusing.
Best regards,
Pavel
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
Download attachment "signature.asc" of type "application/pgp-signature" (196 bytes)
Powered by blists - more mailing lists