Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 11 Apr 2020 14:35:50 -0700
From:      Conrad Meyer <cem@freebsd.org>
To:        Michael Tuexen <tuexen@freebsd.org>
Cc:        src-committers <src-committers@freebsd.org>, svn-src-all <svn-src-all@freebsd.org>,  svn-src-head <svn-src-head@freebsd.org>
Subject:   Re: svn commit: r359809 - head/sys/netinet
Message-ID:  <CAG6CVpVNjZ5PBQxYop-cYb5AbjwNPNJ--dtXoeFC7c2o2sZc0Q@mail.gmail.com>
In-Reply-To: <202004112036.03BKatfm047227@repo.freebsd.org>
References:  <202004112036.03BKatfm047227@repo.freebsd.org>

next in thread | previous in thread | raw e-mail | index | archive | help
Hi Michael,

On Sat, Apr 11, 2020 at 1:37 PM Michael Tuexen <tuexen@freebsd.org> wrote:
>
> Author: tuexen
> Date: Sat Apr 11 20:36:54 2020
> New Revision: 359809
> URL: https://svnweb.freebsd.org/changeset/base/359809
>
> Log:
>   Zero out pointers for consistency.
>
>   This was found by running syzkaller on an INVARIANTS kernel.

For consistency?  If syzkaller found something due to INVARIANTS
sys/queue.h debugging trashing the pointer values, masking them by
writing zeroes doesn't help.  Generally, defeating the kernel's
INVARIANTS system is not wise or useful.

In this use, consider using
'TAILQ_CONCAT(&stcb->asoc.strmout[i].outqueue, &oldstream[i].outqueue,
next)' instead of the loop construct.

Conrad



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAG6CVpVNjZ5PBQxYop-cYb5AbjwNPNJ--dtXoeFC7c2o2sZc0Q>