* Re: [1/7] cpufreq: ap806: Checking implementation ofarmada_8k_cpufreq_init()
[not found] <201904031408372646841@zte.com.cn>
@ 2019-04-03 12:54 ` Markus Elfring
0 siblings, 0 replies; only message in thread
From: Markus Elfring @ 2019-04-03 12:54 UTC (permalink / raw)
To: Wen Yang, Gregory Clement, Viresh Kumar, kernel-janitors
Cc: linux-kernel, linux-arm-kernel, linux-pm, Andrew Lunn,
Jason Cooper, Rafael J. Wysocki, Sebastian Hesselbarth, Yi Wang
>> https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/drivers/cpufreq/armada-8k-cpufreq.c?id=05d08e2995cbe6efdb993482ee0d38a77040861a#n137
>
> Thank you for your comments.
> Adding a null pointer check here is indeed safer.
I wonder then why such a return value check was omitted so far.
> However, the purpose of series of patches is to add missing of_node_put
> to avoid memory leaks.
A few lines might become also more interesting besides these source code places.
> So after that, we will also add a null pointer check later.
Would we like to develop another script for the semantic patch language
to detect similarly questionable implementation details?
Regards,
Markus
^ permalink raw reply [flat|nested] only message in thread