linux-kernel - RE: [PATCH v2] iio: core: Fix IIO_VAL_FRACTIONAL calculation for negative values

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]

Message-ID: <BY5PR02MB6916B943FBBFD6B1733E5C1EA9510@BY5PR02MB6916.namprd02.prod.outlook.com>
Date:   Mon, 31 Aug 2020 08:55:26 +0000
From:   Anand Ashok Dumbre <ANANDASH@...inx.com>
To:     Jonathan Cameron <jic23@...nel.org>
CC:     "knaack.h@....de" <knaack.h@....de>,
        "lars@...afoo.de" <lars@...afoo.de>,
        "pmeerw@...erw.net" <pmeerw@...erw.net>,
        Michal Simek <michals@...inx.com>, git <git@...inx.com>,
        "linux-iio@...r.kernel.org" <linux-iio@...r.kernel.org>,
        "linux-arm-kernel@...ts.infradead.org" 
        <linux-arm-kernel@...ts.infradead.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: RE: [PATCH v2] iio: core: Fix IIO_VAL_FRACTIONAL calculation for
 negative values

Hi Jonathan,

I encountered this when I was developing a new driver.
If you look at the function where this is used, all other IIO_VAL_MICRO and NANO
have this fix added at some point.

Thanks,
Anand

> -----Original Message-----
> From: Jonathan Cameron <jic23@...nel.org>
> Sent: Saturday, August 29, 2020 4:19 PM
> To: Anand Ashok Dumbre <ANANDASH@...inx.com>
> Cc: knaack.h@....de; lars@...afoo.de; pmeerw@...erw.net; Michal
> Simek <michals@...inx.com>; git <git@...inx.com>; linux-
> iio@...r.kernel.org; linux-arm-kernel@...ts.infradead.org; linux-
> kernel@...r.kernel.org; Anand Ashok Dumbre <ANANDASH@...inx.com>
> Subject: Re: [PATCH v2] iio: core: Fix IIO_VAL_FRACTIONAL calculation for
> negative values
> 
> On Wed, 26 Aug 2020 11:14:36 -0700
> Anand Ashok Dumbre <anand.ashok.dumbre@...inx.com> wrote:
> 
> > Fixes IIO_VAL_FRACTIONAL for case when the result is negative and
> > exponent is 0.
> >
> > example: if the result is -0.75, tmp0 will be 0 and tmp1 = 75 This
> > causes the output to lose sign because of %d in snprintf which works
> > for tmp0 <= -1.
> >
> > Signed-off-by: Anand Ashok Dumbre <anand.ashok.dumbre@...inx.com>
> 
> Looks good.  Just one last thing.
> 
> Is this actually hit in an existing driver?  I'm just wondering how far back we
> need to push it in stable etc.
> 
> Thanks,
> 
> Jonathan
> 
> > ---
> > changes since v1:
> > 	Changed -%d to -0 to make the fix clearer.
> > 	Removed the email footer.
> > 	Updated the commit description with an example
> > --
> >  drivers/iio/industrialio-core.c | 8 ++++++--
> >  1 file changed, 6 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/iio/industrialio-core.c
> > b/drivers/iio/industrialio-core.c index cdcd16f1..a239fa2 100644
> > --- a/drivers/iio/industrialio-core.c
> > +++ b/drivers/iio/industrialio-core.c
> > @@ -592,6 +592,7 @@ static ssize_t __iio_format_value(char *buf,
> > size_t len, unsigned int type,  {
> >  	unsigned long long tmp;
> >  	int tmp0, tmp1;
> > +	s64 tmp2;
> >  	bool scale_db = false;
> >
> >  	switch (type) {
> > @@ -614,10 +615,13 @@ static ssize_t __iio_format_value(char *buf,
> size_t len, unsigned int type,
> >  		else
> >  			return scnprintf(buf, len, "%d.%09u", vals[0], vals[1]);
> >  	case IIO_VAL_FRACTIONAL:
> > -		tmp = div_s64((s64)vals[0] * 1000000000LL, vals[1]);
> > +		tmp2 = div_s64((s64)vals[0] * 1000000000LL, vals[1]);
> >  		tmp1 = vals[1];
> >  		tmp0 = (int)div_s64_rem(tmp, 1000000000, &tmp1);
> > -		return scnprintf(buf, len, "%d.%09u", tmp0, abs(tmp1));
> > +		if ((tmp2 < 0) && (tmp0 == 0))
> > +			return snprintf(buf, len, "-0.%09u", abs(tmp1));
> > +		else
> > +			return snprintf(buf, len, "%d.%09u", tmp0,
> abs(tmp1));
> >  	case IIO_VAL_FRACTIONAL_LOG2:
> >  		tmp = shift_right((s64)vals[0] * 1000000000LL, vals[1]);
> >  		tmp0 = (int)div_s64_rem(tmp, 1000000000LL, &tmp1);