Re: Improvement on parallel make

[Brendan Heading]

Philip Guenther writes: 

> I'm also unimpressed by the fact that the cited examples all seem to
> use .WAIT as a kludge around the incomplete dependency tree of a
> recursive make setup.  Making that particular hole easier to live in
> just means more people will dig themselves into it...

I was waiting for someone to say this. 

Is there a use case for this .WAIT syntax outside of helping people not 
understanding how to correctly describe the dependencies for a target ? 
Surely the tool should, where possible, not encourage bad habits. As Philip 
says, surely it would be better to encourage people to lay out their 
dependencies explicitly to enforce the ordering of target builds (which is 
what I always do at present if two targets cannot build in parallel). 

At least for regular C build work, if a build cannot complete properly in 
parallel then the Makefiles almost certainly haven't been written properly. 
I'm sure this is largely true for most other languages out there (though I 
heard a while ago that C++ template stuff can get hairy). 

Brendan