Re: [PATCH] sim: switch to autogenerated ChangeLog files

[Luis Machado <luis.machado@linaro.org>]

On 1/11/21 2:00 PM, Simon Marchi via Gdb-patches wrote:
> On 2021-01-11 6:05 a.m., Andrew Burgess wrote:
>> I think sim/ should follow the same policy as gdb/ for now.  Do feel
>> free to raise this as a suggestion for gdb/ in general though, it
>> would be an interesting conversation to observe.
> 
> Last time I tried the gitlog-to-changelog script on GDB, it produced
> horrible results.  Probably because it was not designed for C++.
> 
> Making a script to produce ChangeLogs for C code is already very
> difficult, making it work with good results for C++ would be even
> worst.  And most importantly, I think it would be wasted development
> time.
> 
> My opinion is that we should either keep hand-written ChangeLogs, or
> admit that ChangeLogs are not useful and get rid of them altogether
> (and you know which side I lean towards).  But auto-generating
> ChangeLogs, other than allowing to check the "we have ChangeLog" box,
> doesn't produce anything useful.

Ever since we moved binutils-gdb to git, ChangeLog's started to feel 
like a waste of time. Now, with broader use of C++, it is even worse. It 
makes things more verbose and harder for tools to find the right scope 
(is the change in a function? A class? A member function?). And the 
limit of 80 columns adds to that burden, since we need to watch out for 
the length of the line. It feels like a puzzle sometimes.

"git log" produces very useful information, and we make the commit 
messages very verbose (more than other projects) so developers can have 
a pretty good idea of what has changed and why.

I'm strongly in favor of dropping ChangeLog's from GDB.