Re: [PATCH] ext4: memory leakage in ext4_mb_init()

[jing zhang <zj.barak@gmail.com>]

2010/4/5, Eric Sandeen <sandeen@redhat.com>:
> jing zhang wrote:
>> 2010/4/5, tytso@mit.edu <tytso@mit.edu>:
>>> On Sun, Apr 04, 2010 at 09:05:14AM +0800, jing zhang wrote:
>>>
>>> How much testing are you doing before submitting patches, out of
>>> curiosity?
>>
>> Yes, Ted, it is curiosity that drives me to do hard works, including patch
>> ext4.
>
> It is the language barrier that is making some of this difficult,
> but I'm not complaining - you speak English much better than I speak any

You are good guy:) Is English your native language? And, I am curious,
what is the second language you are able to speak, Eric?

> second language.  :)
>
> Ted meant that -he- was curious about how much testing you were doing.

How do know what Ted meant, by iphone?

>
> ...
>
>> And after operations on cmdline, I compile the modified, modprobe, dd,
>> and rmmod with virtual machine. It is not hard.
>
> More testing than this would be good; dd is very minimal.
>
> One of our new standard tests for et4 is the xfstests test suite from
> http://git.kernel.org/?p=fs/xfs/xfstests-dev.git;a=summary
>
> It is a collection of many tests developed for xfs, but many tests are
> generic and can run on ext4 as well.  I would suggest that after you have
> several patches ready, you should at least run through the tests in this
> collection.  It won't catch every mistake but it runs a large variety of
> tests, much more stressful than dd.

Thanks for your good advice.

>
> Thanks for your email, and thanks for clearly spending time looking for
> ways to improve ext4.  I think that with practice, you will be a good
> contributor.
>
> Ted can certainly be a patient maintainer - read his suggestions and the
> kernel patch submission guidelines, and I think you will get better at this.
>
> Do your best to explain the reasons for your patches, and any testing you
> have done, and describe any test which can show a bug that you find -
> and we can help to clarify changelogs if they need it.
>

I am not good at testing, partially because it is hard to setup the
required environment, sometimes several hard disks are needed, maybe a
few boxes, but I try to analyse the C code while reading and
understanding the works by great maintainers and developers of Linux
kernel.

                - zj