Re: PATCH libffi win64 support

[Dave Korn <dave dot korn dot cygwin at googlemail dot com>]

Andrew Haley wrote:
> Dave Korn wrote:
>> Ralf Wildenhues wrote:
>>> * Andrew Haley wrote on Tue, Jun 16, 2009 at 08:19:39PM CEST:
>>>> Ralf Wildenhues wrote:
>>>>> autoconf without --force looks at time stamps of input files, yes.
>>>> Oh, no!  That is a classic case of an accident waiting to happen.
>>> Well, it's also a nontrivial speedup in common use cases.
>>>
>>> Cheers,
>>> Ralf
>>   ? I would have thought "running autoconf" doesn't count as a "common"
>> operation but as a rare and almost always explicitly user-directed one.  The
>> risk/benefit tradeoff here seems wrong to me.
> 
> me2.  I was wondering if a patch to change this behaviour to using
> a checkssum rather than a time stamp might be accepted.
> 
> Andrew.

  I'd suggest that the optimisation could be removed altogether;
timestamp-based dependency tracking belongs in make - even though that would
also have been tricked in this case, but then autoconf itself should run
unconditionally, so the manual reconfiguring would have worked.  But I dunno
if maybe the autoconf guys have been down this road before and what the
reasons for implementing it in autoconf itself were when it was done.

    cheers,
      DaveK