Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Wed, 29 Jul 2020 13:52:56 -0700
From:   Kees Cook <keescook@chromium.org>
To:     David Howells <dhowells@redhat.com>
Cc:     peterz@infradead.org, Xiaoming Ni <nixiaoming@huawei.com>,
        David Windsor <dwindsor@gmail.com>,
        Hans Liljestrand <ishkamiel@gmail.com>,
        Elena Reshetova <elena.reshetova@intel.com>,
        Paul Moore <paul@paul-moore.com>, edumazet@google.com,
        paulmck@kernel.org, shakeelb@google.com,
        James Morris <jamorris@linux.microsoft.com>,
        alex.huangjianhui@huawei.com, dylix.dailei@huawei.com,
        chenzefeng2@huawei.com, linux-kernel@vger.kernel.org,
        Will Deacon <will@kernel.org>
Subject: Re: [RFC][PATCH] locking/refcount: Provide __refcount API to obtain
 the old value
Message-ID: <202007291352.8775B08DA@keescook>
References: <20200729111120.GA2638@hirez.programming.kicks-ass.net>
 <20200721195132.GJ10769@hirez.programming.kicks-ass.net>
 <202006142054.C00B3E9C9@keescook>
 <20200612183450.4189588-1-keescook@chromium.org>
 <7be4d56b-0406-099b-e505-02e074c5173e@huawei.com>
 <544539.1595328664@warthog.procyon.org.uk>
 <202007211144.A68C31D@keescook>
 <3211866.1595933798@warthog.procyon.org.uk>
 <4041520.1596055297@warthog.procyon.org.uk>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <4041520.1596055297@warthog.procyon.org.uk>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, Jul 29, 2020 at 09:41:37PM +0100, David Howells wrote:
> peterz@infradead.org wrote:
> 
> > I'm not entirely sure what you mean with interpret, provided you don't
> > trigger a refcount fail, the number will be just what you expect and
> > would get from refcount_read(). If you do trigger a fail, you'll get a
> > negative value.
> 
> That's fine.  I seem to remember talk about the possibility that the number
> wouldn't necessarily bottom out at zero - for instance if it was arranged such
> that the overflow flag was set on an overflow or underflow so that it could be
> trapped on (using INTO or TRAPV, for example).

The trap is an internal detail. The saturation value will be negative,
though.

-- 
Kees Cook