linux-kernel - Re: [PATCH] mmc: sdhci-msm: Fix possible NULL pointer exception

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]

Date:   Tue, 5 Jan 2021 13:20:04 -0600
From:   Bjorn Andersson <bjorn.andersson@...aro.org>
To:     Md Sadre Alam <mdalam@...eaurora.org>
Cc:     adrian.hunter@...el.com, ulf.hansson@...aro.org,
        linux-arm-msm@...r.kernel.org, linux-mmc@...r.kernel.org,
        linux-kernel@...r.kernel.org, sricharan@...eaurora.org
Subject: Re: [PATCH] mmc: sdhci-msm: Fix possible NULL pointer exception

On Tue 22 Dec 02:48 CST 2020, Md Sadre Alam wrote:

> of_device_get_match_data returns NULL when no match.
> So add the NULL pointer check to avoid dereference.
> 
> Signed-off-by: Md Sadre Alam <mdalam@...eaurora.org>
> ---
>  drivers/mmc/host/sdhci-msm.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/mmc/host/sdhci-msm.c b/drivers/mmc/host/sdhci-msm.c
> index 9c7927b..f20e424 100644
> --- a/drivers/mmc/host/sdhci-msm.c
> +++ b/drivers/mmc/host/sdhci-msm.c
> @@ -2235,6 +2235,8 @@ static int sdhci_msm_probe(struct platform_device *pdev)
>  	 * the data associated with the version info.
>  	 */
>  	var_info = of_device_get_match_data(&pdev->dev);
> +	if (!var_info)

To get this you (the SDHCI driver developer) needs to add an entry in
sdhci_msm_dt_match[] without a .data specified.

> +		goto pltfm_free;

And this will cause sdhci_msm_probe() to return 0, giving no hint to
said developer that they screwed up. Given that this can only be caused
by a developer working on this driver you should provide feedback
suitable for such developer, e.g. by:

	if (WARN_ON(!var_info))
		return -EINVAL;

But given that this is only for the developer himself, I don't see that
this adds any value over just reading the callstack you get from the
panic when the next line dereferences var_info (NULL).

Regards,
Bjorn

>  
>  	msm_host->mci_removed = var_info->mci_removed;
>  	msm_host->restore_dll_config = var_info->restore_dll_config;
> -- 
> 2.7.4
>