linux-kernel - Re: [PATCH] iio: vadc: Fix potential dereference of NULL pointer

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: <YoUtezVzsPB8MAEO@google.com>
Date:   Wed, 18 May 2022 10:31:39 -0700
From:   Matthias Kaehlcke <mka@...omium.org>
To:     Yongzhi Liu <lyz_cs@....edu.cn>
Cc:     agross@...nel.org, bjorn.andersson@...aro.org, jic23@...nel.org,
        lars@...afoo.de, linux-arm-msm@...r.kernel.org,
        linux-iio@...r.kernel.org, linux-kernel@...r.kernel.org,
        fuyq@....pku.edu.cn
Subject: Re: [PATCH] iio: vadc: Fix potential dereference of NULL pointer

On Tue, May 17, 2022 at 11:43:00PM -0700, Yongzhi Liu wrote:
> The return value of vadc_get_channel() needs to be checked
> to avoid use of NULL pointer, which is followed by
> the caller 'vadc_do_conversion' of function 'vadc_configure'.
> Fix this by adding the null pointer check on prop
> in function 'vadc_configure'.
> 
> Signed-off-by: Yongzhi Liu <lyz_cs@....edu.cn>
> ---
>  drivers/iio/adc/qcom-spmi-vadc.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/drivers/iio/adc/qcom-spmi-vadc.c b/drivers/iio/adc/qcom-spmi-vadc.c
> index 34202ba..d99bd72 100644
> --- a/drivers/iio/adc/qcom-spmi-vadc.c
> +++ b/drivers/iio/adc/qcom-spmi-vadc.c
> @@ -210,6 +210,9 @@ static int vadc_configure(struct vadc_priv *vadc,
>  	u8 decimation, mode_ctrl;
>  	int ret;
>  
> +	if (!prop)
> +		return -ENODEV;
> +
>  	/* Mode selection */
>  	mode_ctrl = (VADC_OP_MODE_NORMAL << VADC_OP_MODE_SHIFT) |
>  		     VADC_ADC_TRIM_EN | VADC_AMUX_TRIM_EN;


Shouldn't the check be done in vadc_measure_ref_points() where 'prop' is
obtained, rather than deep down in the call chain? For example
vadc_do_conversion() would also dereference the NULL pointer unless one
of the prior function calls fails.