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, 8 Mar 2019 08:41:58 -0400
From:   Jason Gunthorpe <jgg@ziepe.ca>
To:     Rasmus Villemoes <linux@rasmusvillemoes.dk>
Cc:     Bart Van Assche <bvanassche@acm.org>,
        Kees Cook <keescook@chromium.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "linux-rdma@vger.kernel.org" <linux-rdma@vger.kernel.org>,
        Leon Romanovsky <leonro@mellanox.com>
Subject: Re: [PATCH] Avoid that check_shl_overflow() triggers a compiler
 warning when building with W=1
Message-ID: <20190308124158.GD9321@ziepe.ca>
References: <20190307010153.81157-1-bvanassche@acm.org>
 <20190307012417.GU1758@mellanox.com>
 <8a5bd9ae-ebfe-687c-2868-d0f2a610d1e0@acm.org>
 <2c122567-d14b-7867-9230-67f570c13d15@rasmusvillemoes.dk>
 <1552003703.45180.17.camel@acm.org>
 <2d60debb-2b64-1b1f-7e16-d20720c93e28@rasmusvillemoes.dk>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <2d60debb-2b64-1b1f-7e16-d20720c93e28@rasmusvillemoes.dk>
User-Agent: Mutt/1.9.4 (2018-02-28)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Fri, Mar 08, 2019 at 08:58:21AM +0100, Rasmus Villemoes wrote:
> On 08/03/2019 01.08, Bart Van Assche wrote:
> > On Thu, 2019-03-07 at 08:18 +0100, Rasmus Villemoes wrote:
> >> On 07/03/2019 03.14, Bart Van Assche wrote:
> >>> On 3/6/19 5:24 PM, Jason Gunthorpe wrote:
> >>>>>
> >>>>> diff --git a/include/linux/overflow.h b/include/linux/overflow.h
> >>>>> index 40b48e2133cb..8afe0c0ada6f 100644
> >>>>> +++ b/include/linux/overflow.h
> >>>>> @@ -202,6 +202,24 @@
> >>>>>     #endif /* COMPILER_HAS_GENERIC_BUILTIN_OVERFLOW */
> >>>>>   +/*
> >>>>> + * Evaluate a >= 0 without triggering a compiler warning if the type
> >>>>> of a
> >>>>> + * is an unsigned type.
> >>>>> + */
> >>>>> +#define is_positive(a) ({                    \
> >>
> >> is_non_negative, please! positive means > 0. And perhaps it's better to
> >> move these utility macros closer to the top of the file, together with
> >> the other type/range helpers.
> > 
> > Hi Rasmus,
> > 
> > Thank you for the feedback. But according to what I found online opinions
> > about whether or not zero is a positive number seem to vary. From
> > https://en.wikipedia.org/wiki/Sign_(mathematics):
> 
> Yes, I'm a mathematician, I'm aware of that. You can also find people
> who use "less than" in the "<=" sense, and then say "strictly less than"
> when they mean "<".
> 
> > Terminology for signs
> > 
> > When 0 is said to be neither positive nor negative, the following phrases
> > may be used to refer to the sign of a number:
> > * A number is positive if it is greater than zero.
> > * A number is negative if it is less than zero.
> > * A number is non-negative if it is greater than or equal to zero.
> > * A number is non-positive if it is less than or equal to zero.
> > 
> > When 0 is said to be both positive and negative, modified phrases are used
> > to refer to the sign of a number:
> > * A number is strictly positive if it is greater than zero.
> > * A number is strictly negative if it is less than zero.
> > * A number is positive if it is greater than or equal to zero.
> > * A number is negative if it is less than or equal to zero.
> 
> Right, but in no way does it ever make sense to mix these conventions.
> So the options for describing ">= 0, < 0" are "non_negative, negative"
> or "positive, strictly_negative".
> 
> In the context of the C language, the first convention is used. While
> not explicitly stated, it can be inferred from usage of the terms.
> First, the word nonnegative is used (e.g. in defining argc). Second, "If
> the value of the right operand [in a shift expression] is negative [...]
> the behaviour is undefined.", so certainly negative cannot include 0.
> Third, E* constants are required to be positive, and "[errno] is never
> set to zero by any library function". Etc. etc.

Lets use is_unsigned() or is_unsigned_value() then for the name of the
test, that is pretty unambiguous and alot nicer to read than
is_not_negative()

FWIW, in computer science I generally see the terms used as:

  negatve: < 0
  positive: >= 0
  natural: > 0

This language naturally follows the twos complement construction where
it is very logical to say a number with the sign bit set is 'negative'
and a number with it clear is 'positive', which means 0 is positive.

Which is probably enraging to mathematicians.. But has a certain
logic.

.. and most CS places don't actually care about the difference
between > 0 and >= 0 , while < 0 is usually highly interesting.

Jason