In message <[EMAIL PROTECTED]> Robert
Watson writes:
: On Fri, 23 Mar 2001, Warner Losh wrote:
:
: > You'll also get better milage out of make -j N (say 3 or 4) and doing
: > things sequentially. It is safer and runs just as fast.
:
: Dunno if it was a temporary compile problem, but I've actually found that:
:
: make -j 3 buildkernel
:
: hasn't worked properly for me. Either it was a temporary thing and may be
: fixed now, or it's a property of the buildkernel dependencies, and should
: probably be fixed (my kernel build is substantially faster with just a bit
: of parallelism to keep the CPU busy).
I think that it works. I know that the "old way" works with -j values
up to 20 (haven't tried anything higher). There were issues with -j
for a while, but those have been fixed. They were inadvertantly
introduced when we went to building modules.
Warner
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
