linux-kernel - Re: [PATCH] phy: allwinner: Fix GENMASK misuse

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Fri, 8 Nov 2019 09:29:07 +0100
From:   Paul Kocialkowski <paul.kocialkowski@...tlin.com>
To:     Russell King - ARM Linux admin <linux@...linux.org.uk>
Cc:     Rikard Falkeborn <rikard.falkeborn@...il.com>, megous@...ous.com,
        mark.rutland@....com, devicetree@...r.kernel.org, arnd@...db.de,
        gregkh@...uxfoundation.org, linux-kernel@...r.kernel.org,
        mripard@...nel.org, kishon@...com, linux-sunxi@...glegroups.com,
        robh+dt@...nel.org, tglx@...utronix.de, wens@...e.org,
        linux-arm-kernel@...ts.infradead.org, icenowy@...c.io
Subject: Re: [PATCH] phy: allwinner: Fix GENMASK misuse

Hi,

On Thu 07 Nov 19, 23:39, Russell King - ARM Linux admin wrote:
> On Thu, Nov 07, 2019 at 09:46:45PM +0100, Rikard Falkeborn wrote:
> > Arguments are supposed to be ordered high then low.
> > 
> > Signed-off-by: Rikard Falkeborn <rikard.falkeborn@...il.com>
> > ---
> > Spotted while trying to add compile time checks of GENMASK arguments.
> > Patch has only been compile tested.
> 
> My feeling, personally, is that GENMASK() really isn't worth the pain
> it causes.  Can we instead get rid of this thing and just use easier
> to understand and less error-prone hex masks please?

One advantage it has is that is matches the order in which bit fields are
usually given in datasheets, so I personally found that it makes verification
of fields much more straightforward and immediate.

My 2 cents are that it makes sense for hardware registers.

Note that I have recently introduced a SHIFT_AND_MASK_BITS macro[0] for a V4L2
driver, that I (and Mauro) would like to move to linux/bits.h eventually.

> I don't care what anyone else says, personally I'm going to stick with
> using hex masks as I find them way easier to get right first time than
> a problematical opaque macro - and I really don't want the effort of
> finding out that I've got the arguments wrong when I build it.  It's
> just _way_ easier and less error prone to use a hex mask straight off.

I guess it's a matter of personal habit.

[0]: https://git.linuxtv.org/media_tree.git/commit/?id=06eff2150d4db991ca236f3d05a9dc0101475aea

Cheers,

Paul

-- 
Paul Kocialkowski, Bootlin
Embedded Linux and kernel engineering
https://bootlin.com

Download attachment "signature.asc" of type "application/pgp-signature" (489 bytes)