Alan Modra <amodra@gmail.com> writes:

> On Mon, Aug 07, 2023 at 01:07:04PM +0200, Arsen Arsenović via Binutils wrote:
>> bfd/ChangeLog:
>> 
>> 	* aclocal.m4: Regenerate.
>> 
>> gold/ChangeLog:
>> 
>> 	* aclocal.m4: Regenerate.
>> 
>> intl/ChangeLog:
>> 
>> 	* aclocal.m4: Regenerate.
>
> Hmm, these three and the gprofng changes are *not* regeneration,
> because the files don't need regenerating at the start of a patch
> series.

Without this base commit, this diff would be inadvertently merged into
subsequent patches, so I decided to separate it out at the start to get
more self-contained patches.

> What's more, if I apply your patch series then when I regenerate at
> the end I get different generated files..

Yes, I also found this odd.  I couldn't find the cause at the time of
writing.

> Maybe it would be better if changes to generated files are left to a
> final patch that does a bulk regeneration.  It would certainly be
> easier to review.

Hmm, my impression is that it'd be easier to handle these patches if
each commit had accurate generated files.  I can switch to a
regeneration after the fact if so desired.
-- 
Arsen Arsenović