Re: [PATCH v2] ipv4 : igmp : optimize timer modify logic in igmp_mod_timer()

[Jun Zhao <mypopydev@gmail.com>]

On Wed, 2011-11-23 at 19:04 -0500, David Miller wrote:
> From: Jun Zhao <mypopydev@gmail.com>
> Date: Thu, 24 Nov 2011 06:54:45 +0800
> 
> > On Wed, 2011-11-23 at 17:28 -0500, David Miller wrote:
> >> From: Jun Zhao <mypopydev@gmail.com>
> >> Date: Thu, 24 Nov 2011 00:38:42 +0800
> >> 
> >> > When timer is pending and expires less-than-or-equal-to new delay,
> >> > we need not used del_timer()/add_timer().
> >> > 
> >> > Signed-off-by: Jun Zhao <mypopydev@gmail.com>
> >> 
> >> You did not answer Eric's question, why are you optimizing this
> >> less-used code path?
> > 
> > 1). Oh, in the RFC 3376 $5.2, Page 23:
> 
> Then your commit message is terrible.
> 
> Your commit message, one the one hand, talks about optimizing the code.
> 
> Your explanation here talks about RFC conformance.
> 
> Your inconsistencies, and how you ignore important questions posed to
> you like Eric's (until I point it out to you) makes your work
> incredibly irritating to review and process.

I think Eric's means the v1 patch have a obvious bug about lock, I
didn't ignore it. ：（

> 
> Your patch submissions need to be more well formed and your commit
> messages need to explain exactly what your goals are with your change
> and how those goals are being met by the patch you are proposing.
> 
> When we read "optimize timer modify logic" how the heck are we
> supposed to know what this change is actually doing?  Why should we
> think that we actually need your change?  How am we supposed to figure
> out that you are fixing an RFC conformance issue?
> 

I got it, Tks. Maybe the terrible commit message lead to the problem,
I just try to make this function more readability.

Yes, I need to describe the goals more accurately in the commit.

> I'm sorry, this patch submission is junk.  Don't send us junk.
> 

I got it.