On Wed, Dec 19, 2012 at 12:22 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Robert Haas <robertmhaas@gmail.com> writes:
>> On Mon, Dec 17, 2012 at 1:34 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>>> This is definitely not per make's contract, either. I think maybe the
>>> "Don't rebuild the list if only the OBJS have changed" hack in common.mk
>>> is a brick or two shy of a load, but I don't know how to fix that.
>
>> I feel like it's been this way for a while - at least I feel like I've
>> noticed this before. I think there is some inevitable kludginess
>> around having one makefile per subdirectory that leads to these kinds
>> of issues. Maybe we should get rid of all the makefiles under
>> src/backend except for the top-level one and just do everything there.
>
> I mentioned this paper last year, but maybe it's time to start
> taking it seriously:
> http://aegis.sourceforge.net/auug97.pdf
+1 from me. I don't know that just fixing src/backend will do a whole
lot to improve build times in and of itself, but I do think it might
reduce the required amount of alchemy to keep things working.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company