Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Fri, 31 Aug 2018 10:27:14 +0200
From:   Luc Van Oostenryck <luc.vanoostenryck@gmail.com>
To:     Jann Horn <jannh@google.com>
Cc:     Thomas Gleixner <tglx@linutronix.de>, linux-kernel@vger.kernel.org,
        Kees Cook <keescook@chromium.org>, Borislav Petkov <bp@suse.de>
Subject: Re: [PATCH] x86/dumpstack: fix address space casting in
 show_opcodes()
Message-ID: <20180831082712.cbfvv655amizvkfq@ltop.local>
References: <20180830194736.210039-1-jannh@google.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20180830194736.210039-1-jannh@google.com>
User-Agent: NeoMutt/20180622
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, Aug 30, 2018 at 09:47:36PM +0200, Jann Horn wrote:
> I sloppily passed a kernel-typed pointer to __range_not_ok(), and sparse
> doesn't like that.
> Make `prologue` a __user pointer (to protect against accidental
> dereferences) and force-cast it to a kernel pointer when calling
> probe_kernel_read(), which will then immediately force-cast it back to a
> user pointer.

It's a bit sad to have to do this.
__range_not_ok() explicitly requires a __user pointer (I don't know
if there is a good reason for it) but  the real job is done by
__chk_range_not_ok(). Can't you use this later instead?


-- Luc Van Oostenryck