linux-kernel - Re: [PATCH v2] staging: atomisp: move null check to earlier point

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: <20200731083856.GF3703480@smile.fi.intel.com>
Date:   Fri, 31 Jul 2020 11:38:56 +0300
From:   Andy Shevchenko <andy.shevchenko@...il.com>
To:     Cengiz Can <cengiz@...nel.wtf>
Cc:     dan.carpenter@...cle.com, devel@...verdev.osuosl.org,
        gregkh@...uxfoundation.org, linux-kernel@...r.kernel.org,
        linux-media@...r.kernel.org, mchehab@...nel.org,
        sakari.ailus@...ux.intel.com
Subject: Re: [PATCH v2] staging: atomisp: move null check to earlier point

On Fri, Jul 31, 2020 at 01:17:38AM +0300, Cengiz Can wrote:
> `find_gmin_subdev` function that returns a pointer to `struct

func() are referred with name followed by parentheses.

> gmin_subdev` can return NULL.

> In `gmin_v2p8_ctrl` there's a call to this function but the possibility
> of a NULL was not checked before its being dereferenced. ie:

'. ie:' -> ', i.e.:'

> ```

Instead just shift right by two spaces.

> /* Acquired here --------v */
> struct gmin_subdev *gs = find_gmin_subdev(subdev);
> 
> /*  v------Dereferenced here */
> if (gs->v2p8_gpio >= 0) {
>     ...
> }

> ```

Drop this as per above comment.

> To avoid the issue, null check has been moved to an earlier point
> and return semantics has been changed to reflect this exception.

> Please do note that this change introduces a new return value to
> `gmin_v2p8_ctrl`.

This rather should explain better the semantics change, e.g.
"Now the function() refuses to take NULL argument and returns an error. We also
WARN() for sake of the debugging."

> [NEW] - raise a WARN and return -ENODEV if there are no subdevices.
>       - return result of `gpio_request` or `gpio_direction_output`.
>       - return 0 if GPIO is ON.
>       - return results of `regulator_enable` or `regulator_disable`.
>       - according to PMIC type, return result of `axp_regulator_set`
>         or `gmin_i2c_write`.
>       - return -EINVAL if unknown PMIC type.

This has to go after cutter '---' line below.

> Caught-by: Coverity Static Analyzer CID 1465536

Reported-by:

And as discussed previously,
Suggested-by: Mauro ...

> Signed-off-by: Cengiz Can <cengiz@...nel.wtf>

The code looks good enough (WARN() will probably go away when driver gains
better quality).

-- 
With Best Regards,
Andy Shevchenko