|Look for info/paper/experience of parallel make email@example.com (1995-08-21)|
|Re: Look for info/paper/experience of parallel make johnm@soda.CSUA.Berkeley.EDU (1995-08-21)|
|From:||johnm@soda.CSUA.Berkeley.EDU (John D. Mitchell)|
|Organization:||Computer Science Undergrad Assoc., Univ. of Calif. Berkeley|
|Date:||Mon, 21 Aug 1995 23:38:34 GMT|
Pang Hung Hing Anthony <firstname.lastname@example.org> wrote:
> I am looking for information, paper and experience of parallel
>make. I know that some times ago people have talk about this topic, would
>you like to tell me again?
What sort of information are you looking for?
Adam de Boor's PMake has split off into three or four different camps.
Adam works at Geoworks and they use it and have modified it but have not
release their fixes/modifications/additions/etc. back into the public
version. Andreas Stolcke has been maintaining a public version (available
from ftp://ftp.icsi.berkeley.edu/pub/ai/stolcke/software/) which has been
ported to more architectures, etc. He also provides patches to recent
versions of GNU Make to support distributed builds (gmake requires the
'customs' package out of the pmake distribution for this). BSD v4.4 ships
a hacked version of an old version of pmake as the standard make. It seems
to have been gutted and has a fair number of bugs.
GNU Make in its default form can spawn off parallel build jobs but only on
a single machine.
In the last few years, a number of commercial parallel and/or distributed
makes. Symantec came out with one that runs on a Novell Netware network.
Atria's ClearCase product has something too from what I've read but I have
As far as performance goes, it's a definite win if you're running in a
Hope this helps,
Return to the
Search the comp.compilers archives again.