linux-kernel - Re: [PATCH] clocksource/drivers/integrator-ap: add missing of_node

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]

Message-ID: <f36354fc-566a-a3f4-ff01-8aa99d5af753@linaro.org>
Date:   Sun, 25 Nov 2018 10:41:44 +0100
From:   Daniel Lezcano <daniel.lezcano@...aro.org>
To:     Frank Lee <tiny.windzz@...il.com>
Cc:     tglx@...utronix.de, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] clocksource/drivers/integrator-ap: add missing
 of_node_put()

On 25/11/2018 05:25, Frank Lee wrote:
> On Sun, Nov 25, 2018 at 3:49 AM Daniel Lezcano
> <daniel.lezcano@...aro.org> wrote:
>>
>> On 24/11/2018 15:58, Frank Lee wrote:
>>> On Thu, Nov 22, 2018 at 11:23 PM Yangtao Li <tiny.windzz@...il.com> wrote:
>>>>
>>>> of_find_node_by_path() acquires a reference to the node
>>>> returned by it and that reference needs to be dropped by its caller.
>>>> integrator_ap_timer_init_of() doesn't do that, so fix it.
>>>>
>>>> Signed-off-by: Yangtao Li <tiny.windzz@...il.com>
>>>> ---
>>>>  drivers/clocksource/timer-integrator-ap.c | 20 ++++++++++++++------
>>>>  1 file changed, 14 insertions(+), 6 deletions(-)
>>>>
>>>> diff --git a/drivers/clocksource/timer-integrator-ap.c b/drivers/clocksource/timer-integrator-ap.c
>>>> index 76e526f58620..1f04069470bb 100644
>>>> --- a/drivers/clocksource/timer-integrator-ap.c
>>>> +++ b/drivers/clocksource/timer-integrator-ap.c
>>>> @@ -177,7 +177,7 @@ static int __init integrator_ap_timer_init_of(struct device_node *node)
>>>>  {
>>>>         const char *path;
>>>>         void __iomem *base;
>>>> -       int err;
>>>> +       int err,rc = 0;
>>>>         int irq;
>>>>         struct clk *clk;
>>>>         unsigned long rate;
>>>> @@ -210,26 +210,34 @@ static int __init integrator_ap_timer_init_of(struct device_node *node)
>>>>                                 "arm,timer-secondary", &path);
>>>>         if (err) {
>>>>                 pr_warn("Failed to read property\n");
>>>> -               return err;
>>>> +               rc = err;
>>>> +               goto out_put_pri_node;
>>>>         }
>>>>
>>>>
>>>>         sec_node = of_find_node_by_path(path);
>>>>
>>>> -       if (node == pri_node)
>>>> +       if (node == pri_node){
>>>>                 /* The primary timer lacks IRQ, use as clocksource */
>>>> -               return integrator_clocksource_init(rate, base);
>>>> +               rc = integrator_clocksource_init(rate, base);
>>>> +               goto out;
>>>> +       }
>>>>
>>>>         if (node == sec_node) {
>>>>                 /* The secondary timer will drive the clock event */
>>>>                 irq = irq_of_parse_and_map(node, 0);
>>>> -               return integrator_clockevent_init(rate, base, irq);
>>>> +               rc = integrator_clockevent_init(rate, base, irq);
>>>> +               goto out;
>>>>         }
>>>>
>>>>         pr_info("Timer @%p unused\n", base);
>>>>         clk_disable_unprepare(clk);
>>>> +out:
>>>> +       of_node_put(sec_node);
>>>> +out_put_pri_node:
>>>> +       of_node_put(pri_node);
>>>>
>>>> -       return 0;
>>>> +       return rc;
>>>>  }
>>>>
>>>>  TIMER_OF_DECLARE(integrator_ap_timer, "arm,integrator-timer",
>>>> --
>>>> 2.17.0
>>>>
>>> Hi Daniel：
>>>
>>> How about this?
>>
>> Hi,
>>
>> I think there is a simpler fix. The pri_node and the sec_node are used
>> as an identifier to compare against the current node, we can directly
>> drop the refcount after getting the node from path as it is not used as
>> pointer. In addition, a single variable is needed, so we end up with a
>> fix like that.
>>
>>         alias_node = of_find_node_by_path(path);
>>         of_node_put(alias_node);
>>         if (node == alias_node)
>>                 return integrator_clocksource_init(rate, base);
> 
> Daniel,
> 
> OK,I will simplify it like you said.Although I think of_node_put
> should be called
> after we don't use node.

Except I'm missing something, we don't use the alias_node at any time.
The node itself has already a refcount which gives us the guarantee it
is valid.

I agree it can appear a bit strange to put the node right after getting
it but in our case the alias_node is used as an ID, not a pointer, so it
is fine.




-- 
 <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