RE: [PATCH 0/6] Don't use BIT() macro in UAPI headers

From: David Laight <David.Laight@ACULAB.COM>
To: 'Mark Rutland' <mark.rutland@arm.com>,
	Christoph Hellwig <hch@infradead.org>
Cc: Joe Richey <joerichey94@gmail.com>,
	"trivial@kernel.org" <trivial@kernel.org>,
	Joe Richey <joerichey@google.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
	"x86@kernel.org" <x86@kernel.org>,
	"H. Peter Anvin" <hpa@zytor.com>,
	Paolo Bonzini <pbonzini@redhat.com>,
	Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>,
	Mauro Carvalho Chehab <mchehab@kernel.org>,
	Zhangfei Gao <zhangfei.gao@linaro.org>,
	Zhou Wang <wangzhou1@hisilicon.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"kvm@vger.kernel.org" <kvm@vger.kernel.org>,
	"linux-arm-kernel@lists.infradead.org" 
	<linux-arm-kernel@lists.infradead.org>,
	"linux-media@vger.kernel.org" <linux-media@vger.kernel.org>,
	"linux-accelerators@lists.ozlabs.org" 
	<linux-accelerators@lists.ozlabs.org>
Subject: RE: [PATCH 0/6] Don't use BIT() macro in UAPI headers
Date: Mon, 24 May 2021 16:34:39 +0000	[thread overview]
Message-ID: <56cdb86fe8984a94b4a7a8073476d849@AcuMS.aculab.com> (raw)
In-Reply-To: <20210524122901.GH1040@C02TD0UTHF1T.local>

From: Mark Rutland
> Sent: 24 May 2021 13:29
> 
> On Mon, May 24, 2021 at 12:46:26PM +0100, Christoph Hellwig wrote:
> > On Thu, May 20, 2021 at 03:43:37AM -0700, Joe Richey wrote:
> > > This patch series changes all UAPI uses of BIT() to just be open-coded.
> > > However, there really should be a check for this in checkpatch.pl
> > > Currently, the script actually _encourages_ users to use the BIT macro
> > > even if adding things to UAPI.
> >
> > Yes.  In fact it should warn about BIT() in general.  It is totally
> > pointless obsfucation that doesn't even save any typing at all.
> 
> That's not quite true; the point is that if you use BIT() consistently,
> then even when you refer to bits 32 to 63 you won't accidentally
> introduce shifts of more than the width of int, and the definition will
> work equally well for assembly and C, which isn't true if you use `1UL`
> in the definition.
> 
> With that in mind it's shorter and clearer than its functional
> equivalent:
> 
>   BIT(x)
>   (UL(1) << (x))
> 
> So IMO it's preferable to use BIT() generally, or _BITUL() in uapi
> headers.

And then, suddenly the compiler warns about truncation of the
high bits when ~BIT(x) is used to mask a 32bit value on 64bit systems.

Once the C standard committee had decided to change from K&R's
'sign preserving' integer promotions to 'value preserving'
you always lose somewhere.

Personally I prefer hex constants - I can't count bits at all.

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)