linux-kernel - Re: [PATCH v2 2/2] hwmon: Add driver for TI INA232 Current and Power Monitor

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: <dc5380a7-e257-441f-a264-7bcfe193de2c@roeck-us.net>
Date: Tue, 14 Jan 2025 11:23:19 -0800
From: Guenter Roeck <linux@...ck-us.net>
To: Leo Yang <leo.yang.sy0@...il.com>
Cc: jdelvare@...e.com, robh@...nel.org, davem@...emloft.net,
 krzk+dt@...nel.org, conor+dt@...nel.org, Leo-Yang@...ntatw.com,
 corbet@....net, Delphine_CC_Chiu@...ynn.com, linux-hwmon@...r.kernel.org,
 devicetree@...r.kernel.org, linux-kernel@...r.kernel.org,
 linux-doc@...r.kernel.org, kernel test robot <lkp@...el.com>
Subject: Re: [PATCH v2 2/2] hwmon: Add driver for TI INA232 Current and Power
 Monitor

On 1/14/25 00:02, Leo Yang wrote:
> Hi Guenter,
> 
> On Sat, Jan 11, 2025 at 12:22 AM Guenter Roeck <linux@...ck-us.net> wrote:
>>
>>> +
>>> +     /* If INA233 skips current/power, shunt-resistor and current-lsb aren't needed. */
>>> +     /* read rshunt value (uOhm) */
>>> +     if (of_property_read_u32(client->dev.of_node, "shunt-resistor", &rshunt) < 0)
>>> +             rshunt = INA233_RSHUNT_DEFAULT;
>>> +
>>> +     /* read current_lsb value (uA) */
>>> +     if (of_property_read_u16(client->dev.of_node, "ti,current-lsb", &current_lsb) < 0)
>>> +             current_lsb = INA233_CURRENT_LSB_DEFAULT;
>>> +
>> of_property_read_u16() returns -EOVERFLOW if the value provided was too large.
>> This should be checked to avoid situations where the value provided in devicetree
>> is too large.
>>
> Sorry I have a question, I can't get it to return -EOVERFLOW when I test it
> I am using the following properties:
> test16 = /bits/ 16 <0xfffd>;
> of_property_read_u16 reads 0xfffd
> 
> test16o = /bits/ 16 <0xfffdd>;
> of_property_read_u16 reads 0xffdd
> 
> test16o = <0xfffdd>;
> of_property_read_u16 reads 0xf
> 
> test16array = /bits/ 16 <0xfffd 0xfffe>;
> of_property_read_u16 reads 0xfffd
> 
> The same result for device_property_read_u16, it seems that a data
> truncation occurs and none of them return -EOVERFLOW.
> 
> So maybe there is no need to check EOVERFLOW?
> Or maybe we could use the minimum and maximum of the binding to
> indicate the range.
> 

of_property_read_u16() calls of_property_read_u16_array(). The API documentation
of of_property_read_u16_array() lists -ENODATA and -EOVERFLOW as possible error
return values, in addition to -EINVAL. Ignoring those errors is not a good idea,
even if it is not immediately obvious how to reproduce them.

Guenter