linux-kernel - Re: [RFC: sound/i2c/cs8427] clang warns about cast to signed char

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]

Message-ID: <s5hpnw7v9yq.wl-tiwai@suse.de>
Date:   Thu, 18 Oct 2018 12:13:33 +0200
From:   Takashi Iwai <tiwai@...e.de>
To:     "Philipp K" <philipp97kl@...il.com>
Cc:     "Lukas Bulwahn" <lukas.bulwahn@...il.com>,
        <kernelnewbies@...nelnewbies.org>, <perex@...ex.cz>,
        <linux-kernel@...r.kernel.org>
Subject: Re: [RFC: sound/i2c/cs8427] clang warns about cast to signed char

On Thu, 18 Oct 2018 12:10:28 +0200,
Philipp K wrote:
> 
> Hi,
> 
> I tried compiling the cs8247 sound module with clang and hit the
> following warning:
> 
> sound/i2c/cs8427.c:141:31: warning: implicit conversion from 'int' to
> 'char' changes value from 160 to -96 [-Wconstant-conversion]
>         data[0] = CS8427_REG_AUTOINC | CS8427_REG_CORU_DATABUF;
>                 ~ ~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~
> 
> data is a char array, and CS8427_REG_AUTOINC is defined as 0x80 (dec
> 128) which is too big for a (signed) char.
> 
> Looking at other code in the same file, we usually use unsigned chars.
> 
> So is there a reason for data being signed in this case? Otherwise we
> could just change data to unsigned char and be on the safe side.

No any reason I know of.  It must be just an old convention.

> If you want I can send in a patch.

Yes, please.


thanks,

Takashi