linux-kernel - Re: [PATCH] avoid crashing the kernel

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: <YVAYgp+XnCkjPigd@kroah.com>
Date:   Sun, 26 Sep 2021 08:51:46 +0200
From:   Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:     Michael Estner <michaelestner@....de>
Cc:     Lee Jones <lee.jones@...aro.org>, linux-staging@...ts.linux.dev,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] avoid crashing the kernel

On Sat, Sep 25, 2021 at 10:04:30PM +0200, Michael Estner wrote:
> To avoid chrashing the kernel I use WARN_ON instead.
> 
> Signed-off-by: Michael Estner <michaelestner@....de>
> ---
>  drivers/staging/most/i2c/i2c.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)

Hi,

First off, thanks for the changes, but they need a bit more work.

Your subject line should match the others done for this file, so you
need a "staging: most:" prefix like others.  To see this better, do:
	git log --oneline the_file_i_am_touching.c


Also take a look at the kernel documentation for how to write a good
changelog text.  Your wording above needs some work...

> diff --git a/drivers/staging/most/i2c/i2c.c b/drivers/staging/most/i2c/i2c.c
> index 7042f10887bb..e1edd892f9fd 100644
> --- a/drivers/staging/most/i2c/i2c.c
> +++ b/drivers/staging/most/i2c/i2c.c
> @@ -68,7 +68,7 @@ static int configure_channel(struct most_interface *most_iface,
>  	struct hdm_i2c *dev = to_hdm(most_iface);
>  	unsigned int delay, pr;
> 
> -	BUG_ON(ch_idx < 0 || ch_idx >= NUM_CHANNELS);
> +	WARN_ON(ch_idx < 0 || ch_idx >= NUM_CHANNELS);

You really aren't changing much here, for systems running with "panic on
warn", right?

To solve this correctly you should either:
	- determine that these are impossible to hit and remove the test
	  entirely
	- determine that these are possible to hit, and turn them into a
	  real test and handle the error properly.

thanks,

greg k-h