Hi,
Hans-Joachim Lankenau wrote:
[ ... ]
Heck, what machine are you compiling on? :) Recompiling all of, say,
dbaccess or connectivity or svtools or svx or extensions (modules where
I usually work) takes *much* longer (at least on Windows), and they are
smaller than sc.
So: No, I really consider build time an issue here.
(and I'm talking about Windows local builds only. Building a Windows OOo
on a network machine takes *days*! With complete dependencies, this is
more likely to hit us ...)
from my point of view, these valid concerns about complete dependencies
can only be answered by pure build speed. that's what kai is promising.
so the main point to me is to lower the price for an usefull feature so
it gets usable in daily work.
if build times, even with complete dependencies, can be lowered that
dramatically as some of kais data suggest, it's the point to think of
migrating to jam instead of dmake.
>
> so i'm eagerly waiting for the first "jam build" prototype to get my
> hands on ;)
>
I would hate it if complete dependencies were mandatory. And I'm still
skeptical which (jam dependent, precompiled header do not count because
say are doable with dmake, too) mechanism would make compiling so much
faster that we can allow us to compile so much more with full
dependencies. But numbers will convince me :-) - tough I still would
like to know where the speedups originate from.
@Kay: I recently discovered that we use makedepend in a potential
inefficient way. makedepend is designed to be called once per makefile
with all *.cxx in one call. We call it once for every .cxx. I know Ause
will have a look into this. Did you ever play around with this? AFAIK a
lot of the build time on windows is lost in the makedepend phase.
We should also not forget that the dependency generating/checking stuff
is mostly a problem on windows. It's dwarfed by compile time on Unix. So
switching on full dependencies might alienate the Unix users even if
there is a huge win on Windows.
[ ... ]
Heiner
--
Jens-Heiner Rechtien
[EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
