[Simh] A new version...

[Terry Newton]

--- On Fri, 9/19/08, Philipp Hachtmann <hachti at hachti.de> wrote:
> Ok, folks,
> 
> what shall I say? I'm getting frustrated. Spent nearly
> two days with the 
>   Makefile and a suitable basic readline support...
> Yes, Improvements are bad. Dangerous. Political incorrect.
> Sorry.

You're taking it wrong. I was going to retest the
makefile with the string fixes but the link didn't work.
Even if it wasn't compatible enough yet to work on every
platform, it did work quite well under Linux and should
be available for developers to use, regardless of whether
or not it's accepted for the main archive. Testing to
the required standard is a long drawn out process,
dozens if not hundreds of cases have to be determined,
concerns addressed etc. You just work through them,
or do what I do to avoid all that mess, stick it on
a web site and not worry about it. Then people can
use it or not use it... for what I do if only one
person besides me finds it useful, then it's worth
making it available. From what I was reading, some
people found the new makefile useful (I do), or at
least that the more proper approach should be explored
(which I agree with provided it doesn't break it),
so hopefully you're just updating the file.

And in regards to another comment that was posted
concerning my limited virtual XP setup - that setup
can compile 4 instances of simh at once, in the
background while I still use Linux. Not broken.
In fact it works better than real XP with all my ram.

I rebooted into XP using the same MinGW compiler and
got the same exact error but I discovered 2 things...
first when I updated mingw32-make to version 3.81 I got
a different error message, and I had an older gcc 2.95
on my real XP install which of course didn't work,
but that make actually got to try to compile code -
so whatever the problem was (presumably a long string),
it seems to be specific to the MinGW version of make.

Terry