linux-kernel - Re: drivers/hwmon/jc42.c:477 jc42_readable_reg() warn: always true condition '(reg >= 0) => (0-u32max >= 0)'

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] [day] [month] [year] [list]

Date:   Sun, 25 Dec 2022 15:18:27 +0100
From:   Martin Blumenstingl <martin.blumenstingl@...glemail.com>
To:     Guenter Roeck <linux@...ck-us.net>
Cc:     kernel test robot <lkp@...el.com>, oe-kbuild-all@...ts.linux.dev,
        linux-kernel@...r.kernel.org, linux-hwmon@...r.kernel.org
Subject: Re: drivers/hwmon/jc42.c:477 jc42_readable_reg() warn: always true
 condition '(reg >= 0) => (0-u32max >= 0)'

Hi Guenter,

On Fri, Dec 23, 2022 at 12:41 AM Guenter Roeck <linux@...ck-us.net> wrote:
>
> On Thu, Dec 22, 2022 at 10:20:13PM +0100, Martin Blumenstingl wrote:
> > Hi Guenter et al.,
> >
> > On Thu, Dec 22, 2022 at 3:36 PM kernel test robot <lkp@...el.com> wrote:
> > [...]
> > >    475  static bool jc42_readable_reg(struct device *dev, unsigned int reg)
> > >    476  {
> > >  > 477          return (reg >= JC42_REG_CAP && reg <= JC42_REG_DEVICEID) ||
> > >    478                  reg == JC42_REG_SMBUS;
> > The bot is right: we can omit "reg >= JC42_REG_CAP" as it's already
> > covered by the fact that:
> > - the reg variable is unsigned, which means the lower limit is zero
> > - reg <= JC42_REG_DEVICEID covers the upper limit
> >
> > Before I send a patch I'd like to hear if removal of "reg >=
> > JC42_REG_CAP" makes sense to other people.
> >
>
> The bot keeps complaining about it. Yes, it is technically unnecessary,
> but I left it in on purpose to indicate that JC42_REG_CAP is the first
> register and that it wasn't forgotten. Any modern C compiler notices
> that the check is unnecessary and drops it, so there is no runtime penalty.
Thanks for your feedback. Since I had to double check the bot's
complaint I'll just keep this as-is (and not send any patch for this
at all).

> This is one of those situations where I'd like to have a means to tell
> the checker to please stop complaining.
I see, in some cases this may be an actual logic error (for example:
accidentally using an unsigned data type instead of a signed one).


Best regards,
Martin