Re: [PATCH] cpuidle: fix number of initialized/destroyed states

Re: [PATCH] cpuidle: fix number of initialized/destroyed states

[Rafael J. Wysocki]

Hi,

Thanks for the patch!

I'd like Daniel to have a look at it still.

Thanks,
Rafael


On Monday, January 07, 2013 08:12:01 PM Krzysztof Mazur wrote:
> Commit bf4d1b5ddb78f86078ac6ae0415802d5f0c68f92 (cpuidle: support
> multiple drivers) changed the number of initialized state kobjects
> in cpuidle_add_state_sysfs() from device->state_count to drv->state_count,
> but leaved device->state_count in cpuidle_remove_state_sysfs().
> Those two values might have different values, causing for instance
> NULL pointer dereference in cpuidle_remove_state_sysfs().
> 
> Signed-off-by: Krzysztof Mazur <krzysiek@podlesie.net>
> ---
> Hi,
> 
> commit bf4d1b5ddb78f86078ac6ae0415802d5f0c68f92
> (cpuidle: support multiple drivers, merged in v3.8-rc1) causes NULL pointer
> dereference in cpuidle_remove_state_sysfs() when I plug the AC line to my
> laptop. I'm using the acpi_idle cpuidle driver and the C4 state is
> available only on when the system runs from battery. The problem still
> exists in v3.8-rc2 and f243b9b46a22e5790dbbc36f574c2417af49a41.
> 
> I noticed that the commit bf4d1b5ddb78f86078ac6ae0415802d5f0c68f92
> (merged in v3.8-rc1) changed device->state_count to drv->state_count
> in only one of two places, which seems to be incorrect. This patch restores
> device->state_count in both places. It fixes the problem on my system.
> 
> Krzysiek
> 
>  drivers/cpuidle/sysfs.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/cpuidle/sysfs.c b/drivers/cpuidle/sysfs.c
> index 3409429..428754a 100644
> --- a/drivers/cpuidle/sysfs.c
> +++ b/drivers/cpuidle/sysfs.c
> @@ -374,7 +374,7 @@ static int cpuidle_add_state_sysfs(struct cpuidle_device *device)
>  	struct cpuidle_driver *drv = cpuidle_get_cpu_driver(device);
>  
>  	/* state statistics */
> -	for (i = 0; i < drv->state_count; i++) {
> +	for (i = 0; i < device->state_count; i++) {
>  		kobj = kzalloc(sizeof(struct cpuidle_state_kobj), GFP_KERNEL);
>  		if (!kobj)
>  			goto error_state;
> 
-- 
I speak only for myself.
Rafael J. Wysocki, Intel Open Source Technology Center.

Re: [PATCH] cpuidle: fix number of initialized/destroyed states

[Daniel Lezcano]

On 01/07/2013 10:58 PM, Rafael J. Wysocki wrote:
> Hi,
> 
> Thanks for the patch!
> 
> I'd like Daniel to have a look at it still.

I agree with this patch. I was about to send exactly the same.

Thanks Krzysztof for fixing this.


> On Monday, January 07, 2013 08:12:01 PM Krzysztof Mazur wrote:
>> Commit bf4d1b5ddb78f86078ac6ae0415802d5f0c68f92 (cpuidle: support
>> multiple drivers) changed the number of initialized state kobjects
>> in cpuidle_add_state_sysfs() from device->state_count to drv->state_count,
>> but leaved device->state_count in cpuidle_remove_state_sysfs().
>> Those two values might have different values, causing for instance
>> NULL pointer dereference in cpuidle_remove_state_sysfs().
>>
>> Signed-off-by: Krzysztof Mazur <krzysiek@podlesie.net>

Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>


-- 
 <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog

Re: [PATCH] cpuidle: fix number of initialized/destroyed states

[Rafael J. Wysocki]

On Monday, January 07, 2013 11:56:36 PM Daniel Lezcano wrote:
> On 01/07/2013 10:58 PM, Rafael J. Wysocki wrote:
> > Hi,
> > 
> > Thanks for the patch!
> > 
> > I'd like Daniel to have a look at it still.
> 
> I agree with this patch. I was about to send exactly the same.
> 
> Thanks Krzysztof for fixing this.
> 
> 
> > On Monday, January 07, 2013 08:12:01 PM Krzysztof Mazur wrote:
> >> Commit bf4d1b5ddb78f86078ac6ae0415802d5f0c68f92 (cpuidle: support
> >> multiple drivers) changed the number of initialized state kobjects
> >> in cpuidle_add_state_sysfs() from device->state_count to drv->state_count,
> >> but leaved device->state_count in cpuidle_remove_state_sysfs().
> >> Those two values might have different values, causing for instance
> >> NULL pointer dereference in cpuidle_remove_state_sysfs().
> >>
> >> Signed-off-by: Krzysztof Mazur <krzysiek@podlesie.net>
> 
> Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>

OK, thanks.

I have a pull request for 3.8-rc3 ready, so I'll send this one in the next
round.

Thanks,
Rafael


-- 
I speak only for myself.
Rafael J. Wysocki, Intel Open Source Technology Center.