linux-kernel - Re: [PATCH 3/3][V3] iio: Handle enumerated properties with gaps

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 for Android: free password hash cracker in your pocket

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

Date:   Thu, 9 May 2019 07:31:27 +0000
From:   "Ardelean, Alexandru" <alexandru.Ardelean@...log.com>
To:     "andriy.shevchenko@...ux.intel.com" 
        <andriy.shevchenko@...ux.intel.com>
CC:     "linux-scsi@...r.kernel.org" <linux-scsi@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "lars@...afoo.de" <lars@...afoo.de>,
        "linux-iio@...r.kernel.org" <linux-iio@...r.kernel.org>,
        "gregkh@...uxfoundation.org" <gregkh@...uxfoundation.org>,
        "jic23@...nel.org" <jic23@...nel.org>
Subject: Re: [PATCH 3/3][V3] iio: Handle enumerated properties with gaps

On Wed, 2019-05-08 at 16:17 +0300, Andy Shevchenko wrote:
> [External]
> 
> 
> On Wed, May 08, 2019 at 02:19:13PM +0300, Alexandru Ardelean wrote:
> > From: Lars-Peter Clausen <lars@...afoo.de>
> > 
> > Some enums might have gaps or reserved values in the middle of their
> > value
> > range. E.g. consider a 2-bit enum where the values 0, 1 and 3 have a
> > meaning, but 2 is a reserved value and can not be used.
> > 
> > Add support for such enums to the IIO enum helper functions. A reserved
> > values is marked by setting its entry in the items array to NULL rather
> > than the normal descriptive string value.
> > 
> > Also, `__sysfs_match_string()` now supports NULL gaps, so that doesn't
> > require any changes.
> > -     for (i = 0; i < e->num_items; ++i)
> > +     for (i = 0; i < e->num_items; ++i) {
> > +             if (!e->items[i])
> > +                     continue;
> >               len += scnprintf(buf + len, PAGE_SIZE - len, "%s ", e-
> > >items[i]);
> > +     }
> 
> The problem here that the user will have no clue where the gap is
> happened, to
> solve this we need either bitmap of array, where set bits shows defined
> items,
> or use comma-separated list of values. The latter would need another node
> since
> we don't break user space.

Hmmm.
I am wondering if there are cases where userspace would care about reserved
values and/or positions of reserved bit-fields.
Maybe you could offer examples/use-cases where this is needed.

To some extent the kernel [drivers & frameworks] should probably not need
to expose that "string-enum-X"  == `bitfield_2` matching; otherwise it
doesn't really add much value ; the whole point of frameworks [in general]
is to offer some level of abstraction to HW.

The only example I can think of [atm], is when a reserved bit-field will be
used in the future. But then, the driver should care about this, and not
the framework. The driver should decide that "bitfield_2" will
enable/disable something [in the future], and should be considered in a
such a way (when being written). If the driver can't make this prediction [
about "bitfield_2"] then a new driver must be written anyway.

But I will agree that I may not have all arguments in mind to be 100% sure
of all this.

Thanks
Alex

> 
> --
> With Best Regards,
> Andy Shevchenko
> 
>