linux-kernel - Re: [PATCH] counter: add defaults to switch-statements

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:   Mon, 28 Feb 2022 08:03:18 -0500
From:   William Breathitt Gray <vilhelm.gray@...il.com>
To:     Jonathan Cameron <jic23@...nel.org>, trix@...hat.com
Cc:     nathan@...nel.org, ndesaulniers@...gle.com,
        Jonathan.Cameron@...wei.com, linux-iio@...r.kernel.org,
        linux-kernel@...r.kernel.org, llvm@...ts.linux.dev
Subject: Re: [PATCH] counter: add defaults to switch-statements

On Sun, Feb 27, 2022 at 05:46:10PM +0000, Jonathan Cameron wrote:
> On Sun, 27 Feb 2022 08:17:46 -0800
> trix@...hat.com wrote:
> 
> > From: Tom Rix <trix@...hat.com>
> > 
> > Clang static analysis reports this representative problem
> > counter-chrdev.c:482:3: warning: Undefined or garbage value
> >   returned to caller
> >   return ret;
> >   ^~~~~~~~~~
> > 
> > counter_get_data() has a multilevel switches, some without
> > defaults, so ret is sometimes not set.
> > Add returning -EINVAL similar to other defaults.
> > 
> > Fixes: b6c50affda59 ("counter: Add character device interface")
> > Signed-off-by: Tom Rix <trix@...hat.com>
> I'm fairly sure this one is warning supression rather than a fix as
> that type u8 being used in the switch is storing an enum which is
> set only by kernel code and all values of the enum are covered.
> 
> However, that's not locally visible so to me the addition looks good, I'd
> just be tempted to drop the fixes tag as backporting this looks like noise
> to me.
> 
> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@...wei.com>

Yes, this isn't really a bug because the only values available are those
from the enum, but I see how this is not immediately obvious at first.
If Tom has no objections, I'll pick this up and drop the fixes tag so
that we have it along with the other counter changes for the 5.18 merge.

Thanks,

William Breathitt Gray

> > ---
> >  drivers/counter/counter-chrdev.c | 4 ++++
> >  1 file changed, 4 insertions(+)
> > 
> > diff --git a/drivers/counter/counter-chrdev.c b/drivers/counter/counter-chrdev.c
> > index b7c62f957a6a8..69d340be9c93f 100644
> > --- a/drivers/counter/counter-chrdev.c
> > +++ b/drivers/counter/counter-chrdev.c
> > @@ -477,6 +477,8 @@ static int counter_get_data(struct counter_device *const counter,
> >  		case COUNTER_SCOPE_COUNT:
> >  			ret = comp->count_u8_read(counter, parent, &value_u8);
> >  			break;
> > +		default:
> > +			return -EINVAL;
> >  		}
> >  		*value = value_u8;
> >  		return ret;
> > @@ -496,6 +498,8 @@ static int counter_get_data(struct counter_device *const counter,
> >  		case COUNTER_SCOPE_COUNT:
> >  			ret = comp->count_u32_read(counter, parent, &value_u32);
> >  			break;
> > +		default:
> > +			return -EINVAL;
> >  		}
> >  		*value = value_u32;
> >  		return ret;
>