Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Mon, 6 Jan 2020 11:18:11 +0530
From:   Viresh Kumar <viresh.kumar@linaro.org>
To:     qiwuchen55@gmail.com
Cc:     kgene@kernel.org, krzk@kernel.org, rjw@rjwysocki.net,
        linux-arm-kernel@lists.infradead.org,
        linux-samsung-soc@vger.kernel.org, linux-pm@vger.kernel.org,
        linux-kernel@vger.kernel.org, chenqiwu <chenqiwu@xiaomi.com>
Subject: Re: [PATCH] cpufreq: s3c: avoid use after free issue in
 xxx_cpufreq_reboot_notifier_evt()
Message-ID: <20200106054811.uduf2qdn5ecvbwrc@vireshk-i7>
References: <1577515439-14477-1-git-send-email-qiwuchen55@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1577515439-14477-1-git-send-email-qiwuchen55@gmail.com>
User-Agent: NeoMutt/20180716-391-311a52
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 28-12-19, 14:43, qiwuchen55@gmail.com wrote:
> From: chenqiwu <chenqiwu@xiaomi.com>
> 
> There is a potential UAF issue in xxx_cpufreq_reboot_notifier_evt() that
> the cpufreq policy of cpu0 has been released before using it. So we should
> make a judgement to avoid it.

There is no UAF problem here, but that we do cpufreq_cpu_get() with a
corresponding cpufreq_cpu_put().

> Signed-off-by: chenqiwu <chenqiwu@xiaomi.com>
> ---
>  drivers/cpufreq/s3c2416-cpufreq.c | 11 ++++++++++-
>  drivers/cpufreq/s5pv210-cpufreq.c | 10 +++++++++-
>  2 files changed, 19 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/cpufreq/s3c2416-cpufreq.c b/drivers/cpufreq/s3c2416-cpufreq.c
> index 1069103..0f576ba 100644
> --- a/drivers/cpufreq/s3c2416-cpufreq.c
> +++ b/drivers/cpufreq/s3c2416-cpufreq.c
> @@ -304,6 +304,7 @@ static int s3c2416_cpufreq_reboot_notifier_evt(struct notifier_block *this,
>  {
>  	struct s3c2416_data *s3c_freq = &s3c2416_cpufreq;
>  	int ret;
> +	struct cpufreq_policy policy;
>  
>  	mutex_lock(&cpufreq_lock);
>  
> @@ -318,7 +319,15 @@ static int s3c2416_cpufreq_reboot_notifier_evt(struct notifier_block *this,
>  	 */
>  	if (s3c_freq->is_dvs) {
>  		pr_debug("cpufreq: leave dvs on reboot\n");
> -		ret = cpufreq_driver_target(cpufreq_cpu_get(0), FREQ_SLEEP, 0);
> +
> +		memset(&policy, 0, sizeof(policy));
> +		ret = cpufreq_get_policy(&policy, 0);
> +		if (ret < 0) {
> +			pr_debug("cpufreq: get no policy for cpu0\n");
> +			return NOTIFY_BAD;
> +		}
> +

This doesn't make sense to me, why don't you do cpufreq_cpu_get() and
put() instead ?

> +		ret = cpufreq_driver_target(&policy, FREQ_SLEEP, 0);
>  		if (ret < 0)
>  			return NOTIFY_BAD;
>  	}
> diff --git a/drivers/cpufreq/s5pv210-cpufreq.c b/drivers/cpufreq/s5pv210-cpufreq.c
> index 5d10030..d99b4b1 100644
> --- a/drivers/cpufreq/s5pv210-cpufreq.c
> +++ b/drivers/cpufreq/s5pv210-cpufreq.c
> @@ -555,8 +555,16 @@ static int s5pv210_cpufreq_reboot_notifier_event(struct notifier_block *this,
>  						 unsigned long event, void *ptr)
>  {
>  	int ret;
> +	struct cpufreq_policy *policy;
>  
> -	ret = cpufreq_driver_target(cpufreq_cpu_get(0), SLEEP_FREQ, 0);
> +	policy = cpufreq_cpu_get(0);
> +	if (!policy) {
> +		pr_debug("cpufreq: get no policy for cpu0\n");
> +		return NOTIFY_BAD;
> +	}
> +
> +	ret = cpufreq_driver_target(policy, SLEEP_FREQ, 0);
> +	cpufreq_cpu_put(policy);

Like what is done here.

Also add a blank line here.

>  	if (ret < 0)
>  		return NOTIFY_BAD;
>  
> -- 
> 1.9.1

-- 
viresh