Re: trimming build/installworld
SUBDIR_CHANGE provides
1) A top level switch which enables/disables the subdir{add|drop} extensions
2) Specifies a directory to define the context for this particular build
variant.
IMHO, the concept needs to be used more.
make.conf is a GLOBAL definition of parameters and like /usr/include should
apply ONLY to the host system. It should not be used to control builds of
alternate systems.
Using a bunch of command line flags is extremely awkward and eventually leads
to namespace conflict problems.
--
Richard Wackerbarth
[EMAIL PROTECTED]
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message
Re: Stability and versions - was Re: Let 3.x die ASAP?
On Fri, 31 Mar 2000, J McKitrick wrote: > "STABLE" refers to the code base, NOT the stability of systems running > it. > Simple concept, deep meaning. Newbies should understand ... And therein lies the problem. Newbies don't understand much of anything about this (or any other) project. The approach with the "common English", (or French, German, etc.) point of view. If they, for whatever reason, select an inappropriate branch, they are very likely to conclude that FreeBSD is a "toy" or otherwise non-serious project and turn their attention to another system. This further hurts FreeBSD because they will also spread word of their poor experience to others. As has often been said, "You only get one chance to make a good first impression". Unless the developers wake up and realize that "market share" is important and take steps to improve it, FreeBSD will continue to be a "back water" inconsequential sand box. We desperately need more companies to consider FreeBSD as their first choice for corporate servers. Market share is significantly affected by first impressions. We need to do EVERYTHING possible to make that first experience a good one -- Especially when it doesn't affect the ultimate performance of the system. Quit thinking of FreeBSD from the developer's point of view. Think of it from the new user's viewpoint. Give them what they expect. The developers are better capable of adapting. And any "fossils" who want to argue "we've always done it that way" need to move over. OS's today are not what they were five, much less thirty, years ago. If you want things like they used to be, stick with BSD 4.4 (if you can still find any hardware to run it) and program in COBOL. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-stable" in the body of the message
Re: slow connections over ethernet
On Sun, 09 Apr 2000, Brian Beattie wrote: > I'm running a 10Mb ethernet and recently it has started taking a very logn > time to connect and log into one system from another. >Once logged in, every thing seems fine. Sounds like inverse DNS lookup delays. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-stable" in the body of the message
Re: HEADS UP! Always use the 'make buildkernel' target to make yer kernels
On Mon, 10 Jul 2000, Greg Lehey wrote: > Agreed. I tried it out and found a number of things I didn't like > about it. Basically, it's a completely different build process: > > 1. Before building, it removes the existing kernel build tree. > There's no good reason for this. Agreed > 2. It builds in a different tree (/usr/obj instead of > /usr/src/sys/compile). These two points mean that if you later > want to go back and tune your kernel (change a driver parameter, > say), you can't just do a config; cd ../../compile/FOO; make, you > have to go the whole nine yards. I've argued (to no avail) that the whole kernel build was always "wrong" and that kernels should be treated in the same structure that is used to build userland programs. In particular, each kernel should get its own source directory just like each each user program does. The kernel has one primary source file. This file gets "compiled" by "config" (just as "foo.y") would get processed by "yacc" before calling "gcc" to actually do the compile. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-stable" in the body of the message
