[Q] Sending a large patch

[Q] Sending a large patch

[Masaru Nomura]

Hi,

Can anyone help me with “how to properly deal with a large patch” please?

Now, I'm cleaning up a file and it has over 100 warnings of line over
80 character issues. As the problem seems to be the same, I may
correct the bugs and send one commit. But I think it's a pain for
reviewers to check all of them within one patch.

What do you think I could do?

Thank you in advance,
Masaru

Re: [Q] Sending a large patch

[Dan Carpenter]

On Wed, May 14, 2014 at 01:48:48AM +0100, Masaru Nomura wrote:
> Hi,
> 
> Can anyone help me with “how to properly deal with a large patch” please?
> 
> Now, I'm cleaning up a file and it has over 100 warnings of line over
> 80 character issues. As the problem seems to be the same, I may
> correct the bugs and send one commit. But I think it's a pain for
> reviewers to check all of them within one patch.
> 
> What do you think I could do?

100 is not bad.  Is this is all in one driver?  I say go ahead and send
it.

regards,
dan carpenter

Re: [Q] Sending a large patch

[Bernd Petrovitsch]

On Mit, 2014-05-14 at 01:48 +0100, Masaru Nomura wrote:
[...]
> Now, I'm cleaning up a file and it has over 100 warnings of line over
> 80 character issues. As the problem seems to be the same, I may

And you are really sure that is more readable afterwards with the
additional line breaks?

> correct the bugs and send one commit. But I think it's a pain for
> reviewers to check all of them within one patch.

Apart from the questionable usefulness of the above, you should split it
per subsystem/maintainer so that people do not need to look for "their"
parts.
Additionally, the maintainer may take the patch (and you are done) or
reject it (and you are also done;-).

	Bernd
-- 
Bernd Petrovitsch                  Email : bernd@petrovitsch.priv.at
                     LUGA : http://www.luga.at

Re: [Q] Sending a large patch

[Masaru Nomura]

Thank you for your reply.

>> Now, I'm cleaning up a file and it has over 100 warnings of line over
>> 80 character issues. As the problem seems to be the same, I may
>
> And you are really sure that is more readable afterwards with the
> additional line breaks?

Well, my intention is to follow the coding style[1].
As far as I understand, we follow this for linux-code readability.
(Of course, I think I shouldn't split/break a line in a bad way though.)

Chap 2 says:
The limit on the length of lines is 80 columns and this is a strongly
preferred limit.

Or are there any situations where we ignore over 80 characters?

>> correct the bugs and send one commit. But I think it's a pain for
>> reviewers to check all of them within one patch.
>
> Apart from the questionable usefulness of the above, you should split it
> per subsystem/maintainer so that people do not need to look for "their"
> parts.

I'm working on just one file so I think the relevant maintainers are the
same. (Do I have it correct?)

Or do you mean I should make multiple patches depending on the information
of who takes responsibility of which part within the same file? If
this is the case,
is there a good tool to do this easily?

Thank you,
Masaru

[1] https://www.kernel.org/doc/Documentation/CodingStyle

Re: [Q] Sending a large patch

[Dan Carpenter]

Just send the patch and we can discuss it from there.

regards,
dan carpenter

Re: [Q] Sending a large patch

[Masaru Nomura]

> Just send the patch and we can discuss it from there.
Sure. I'll try to finish it soon.

Thank you,
Masaru

Re: [Q] Sending a large patch

[Bernd Petrovitsch]

On Mit, 2014-05-14 at 15:07 +0100, Masaru Nomura wrote:
> >> Now, I'm cleaning up a file and it has over 100 warnings of line over
> >> 80 character issues. As the problem seems to be the same, I may
> >
> > And you are really sure that is more readable afterwards with the
> > additional line breaks?
> 
> Well, my intention is to follow the coding style[1].

Of course.

> As far as I understand, we follow this for linux-code readability.
> (Of course, I think I shouldn't split/break a line in a bad way though.)
>
> Chap 2 says:
> The limit on the length of lines is 80 columns and this is a strongly
> preferred limit.
>
> Or are there any situations where we ignore over 80 characters?

There are maintainers who do not see it that strict.
(Probably) no one has problem if one reorganizes 130 character lines. 
But if a line with just some printk() has 81 characters if doesn't help
readability/understandability if it is split just for the sake of the
limit.

	Bernd
-- 
Bernd Petrovitsch                  Email : bernd@petrovitsch.priv.at
                     LUGA : http://www.luga.at