Re: make: should -j affect cwd?

[Mouse <mouse%Rodents-Montreal.ORG@localhost>]

> EHLO Mouse,

250 jkl

>> So, it sounds to me as though a first step would be to document
>> exactly what we have.  Warts and all.
> I would like to agree with you, Mouse, but the existing man page is
> the result of exactly that effort to date.

Oh. :(  I'd assumed it was more like most manpages, documenting a
somewhat indeterminate subset of what various people thought at various
times it was supposed to do.

> If building NetBSD is the minimum goal (and it probably should be) I
> would suggest an ever-receding hack to achieve backward
> compatibility: execute the old code.

This actually makes a good deal of sense to me.

> The first thing is to see if there's agreement on the problem.

Well, I think there's agreement there _is_ a problem; at least, I
haven't seen anyone argue with that proposition.  I'm not entirely sure
what it is, but I'm inclined to say it's a combination of two things:
(1) an incompletely understood and even-more-incompletely documented
program which, despite being very central to practically every software
build, is in the running for, what was it? the ugliest thing not
walking on two legs? and (2) a large collection of driving files for
that program which exercise, nay, depend on, an indeterminate subset of
the functionality (including "functionality" due to bugs at various
levels) of program (1).

My own impression is that make has grown by accretion and is well past
the point where it should have been scrapped and rewritten, but is only
now getting to the point where a rewrite is less pain than adding the
next tweak to the existing blob.

/~\ The ASCII                             Mouse
\ / Ribbon Campaign
 X  Against HTML                mouse%rodents-montreal.org@localhost
/ \ Email!           7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B