linux-kernel - Re: [PATCH 2/3] thermal: core: fix: Initialize the max

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: <20141002222649.GB11510@developer>
Date:	Thu, 2 Oct 2014 18:26:51 -0400
From:	Eduardo Valentin <edubezval@...il.com>
To:	Lukasz Majewski <l.majewski@...sung.com>
Cc:	Zhang Rui <rui.zhang@...el.com>,
	Linux PM list <linux-pm@...r.kernel.org>,
	Lukasz Majewski <l.majewski@...ess.pl>,
	Bartlomiej Zolnierkiewicz <b.zolnierkie@...sung.com>,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/3] thermal: core: fix: Initialize the max_state
 variable to 0

Hello Lukasz,

On Wed, Sep 24, 2014 at 10:27:11AM +0200, Lukasz Majewski wrote:
> Pointer to the uninitialized max_state variable is passed to get the
> maximal cooling state.
> For CPU cooling device (cpu_cooling.c) the cpufreq_get_max_state() is
> called, which even when error occurs will return the max_state variable
> unchanged.
> Since error for ->get_max_state() is not checked, the automatically


Good that you added a fix in your series for this.


> allocated value of max_state is used for (upper > max_state) comparison.
> For any possible max_state value it is very unlikely that it will be less
> than upper.
> As a consequence, the cooling device is bind even without the backed
> cpufreq table initialized.
> 
> This initialization will prevent from accidental binding trip points to
> cpu freq cooling frequencies when cpufreq driver itself is not yet fully
> initialized.

Although I agree with the fix, as long as we also include a check for
the .get_max_state return value,  I believe the problem you are
describing is about initialization sequence.

In general, I believe we need a better
sequencing between thermal and cpufreq subsystems. One way out is to
include a check for cpufreq driver in the thermal driver, and return
-EPROBE_DEFER  when cpufreq is not ready.

> 
> Signed-off-by: Lukasz Majewski <l.majewski@...sung.com>
> ---
>  drivers/thermal/thermal_core.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/thermal/thermal_core.c b/drivers/thermal/thermal_core.c
> index 454884a..747618a 100644
> --- a/drivers/thermal/thermal_core.c
> +++ b/drivers/thermal/thermal_core.c
> @@ -927,7 +927,7 @@ int thermal_zone_bind_cooling_device(struct thermal_zone_device *tz,
>  	struct thermal_instance *pos;
>  	struct thermal_zone_device *pos1;
>  	struct thermal_cooling_device *pos2;
> -	unsigned long max_state;
> +	unsigned long max_state = 0;
>  	int result;
>  
>  	if (trip >= tz->trips || (trip < 0 && trip != THERMAL_TRIPS_NONE))
> -- 
> 2.0.0.rc2
> 

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/