Re: improvements for parallel makes ?

[Alexey Neyman]

Hi Paul,

On Thursday 20 April 2006 17:02, Paul D. Smith wrote:
> I need to read your email in more detail to understand the issue and
> think about possible solutions.  Before that happens I can't say 
> whether or not the method you describe, which seems at first glance
> like a big change both to the code and to some current behaviors,
> will be accepted/acceptable.
> 
> So, if you would like to proceed with this work as a 
> proof-of-concept that's great, just so long as expectations are set
> properly. Alternatively we can continue to have a conversation about
> it and maybe come up with an alternative... or not.
> 
> I'm bogged down with some other things right now so it will take me 
> a week or so to spend the time necessary to think about this issue.

I'm fine with the delay, as I'm leaving tomorrow for a two-week-long 
vacation :) It was just that I was refreshing that patch to apply to 
the contemporary make sources and discovered the misbehavior I 
described. So, I have performed some investigation on how BSD 
implements .WAIT. 

That said, I can start a proof-of-concept work as soon as I return 
(May 8).
 
> There is also a thread on a .WAIT proposal from last year on one of 
> the make mailing lists which discusses a number of scenarios.  You
> may have already read that one.

Yes, it's me who started that thread a year ago :)

Regards,
Alexey.

PS. Paul, I started getting the GNU paperwork done, just in case. They 
sent me a questionnaire, I filled it in and sent it back. There were 
no reply from them since then. I tried to ping them about that, but 
got no reply as well. Is it normal?

-- 
You people are so beautiful... so unspoiled.
                        -- Arilou, SC2