Re: s390/debug: Fine-tuning for several function implementations

From: SF Markus Elfring
Date: Fri Sep 09 2016 - 12:50:43 EST


> While I agree that the old code in arch/s390/kernel/debug.c does not abide to
> the current coding style standards,

Thanks for this kind of acknowledgement.

Is such an information worth for further development considerations?


> I doubt there is much value in these patches.

I assume that your doubts could be adjusted, couldn't they?

Are there any other concerns involved in the background?


> To be honest I got annoyed after the third patch

Which of the proposed changes did trigger such a reaction?


I find this response also a bit surprising because of the aspect
that I offered you some results from my work as a free software developer.


> and stopped reading after the forth.

I imagine that you could have aborted the review of my update suggestions
a bit too early for your debug software module.

I agree that the value is varying for the presented 17 update steps.
But I hope that their value is potentially bigger overall
than you categorise them at first glance.

Now I would like to try to get a bit of your software development attention
once more for two of them at least. I hope that it can be easier to clarify
their value.

* Do the implementations of the functions "debug_areas_alloc"
and "debug_get_user_string" need another look together with a more detailed
source code review?

* How do you think about to use functions like "kmalloc_array"
and "memdup_user" there instead?

Regards,
Markus