Re: [PATCH 3/5] staging: erofs: decrease the shrink count in erofs_workgroup_get

From: Gao Xiang
Date: Wed Jan 16 2019 - 08:02:04 EST


Hi Dan,

On 2019/1/16 18:45, Dan Carpenter wrote:
> On Wed, Jan 16, 2019 at 04:59:54PM +0800, Gao Xiang wrote:
>> It is more suitable to update in erofs_workgroup_get since
>> it's actually the one matched with erofs_workgroup_put.
>>
>
> This patch is fine. No need to resend.
>
> Reviewed-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx>
>
> But for future reference, I found the commit message a bit confusing.
> Ideally, I would understand basically what the commit does and why
> without needing to read the diff.
>
> I mostly just read the subject or the commit message, either or, instead
> of reading both. (This is not really true.) My email client looks like
> this. (Also not true).
>
> https://marc.info/?l=linux-driver-devel&m=154762932712086&w=2
>
> While none of that was strictly true, it is a little bit true-ish...
> So please, assume that some people will start reading the commit message
> without reading the subject.
>
> I sort of thought from reading the commit message that it was a bugfix
> or a behavior change. A better commit message would be:
>
> staging: erofs: move shrink accounting inside the function
>
> This patch moves the &erofs_global_shrink_cnt accounting from the caller
> to erofs_workgroup_get(). It's cleaner and it matches erofs_workgroup_put()
> better. No behavior change.

Thanks for your kindly suggestion.
Actually I have to think more about how to express in English
as a foreign speaker...

But you are absolutely right ;) this commit is actually confusing and I will
do my best to write commit subject and message clearly...

If something wrong like this, please kindly point out and I will update immediately :)

Anyway, I will send patch v2 of this commit for Greg to merge...

Thanks,
Gao Xiang

>
> regards,
> dan carpenter
>