M.Canales.es wrote:
><snip>
> I prefer add it into the optimize code but as a separate setting and maybe
> using a separate function call.
If it's a separate setting and function call, it wouldn't really be in
the optimize code, right? ;)
I figured something simple like the following called after wrt_optimize:
---
wrt_makeflags() {
if '-j n' set (or MAKEFLAGS=n defined in common/config),
echo "MAKEFLAGS=$n" to Makefile, where n is the number of jobs.
}
---
I think that's more or less what you're getting at, except I don't see
the point of putting the MAKEFLAGS setting anywhere but in common/config.
> It don't optimize the binaries, but optimize the build time.
That was kinda my point; it's not the binaries but jhalfs that is being
affected. Knowing this, why mix a jhalfs behavior-related variable in
with package optimizations. MAKEFLAGS doesn't really optimize anything
in the same sense as the wrt_optimize() stuff, it just tells jhalfs to
do more at once.
>> I'm still in the process of running the optimization tests; I set it to
>> run 3 builds storing the reports and leave the last build with "-O 1"
>> (for logs).
>
> Well, the ICA/farce report for that last iteration should to show that almost
> all binaries differ ;-)
Didn't run it with ICA/farse. I'll get to that shortly.
There doesn't seem to be much difference in build time whether using
optimizations or not.
Regards,
Jeremy.
--
http://linuxfromscratch.org/mailman/listinfo/alfs-discuss
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page
