PureBasic Forums - English

When I attempt to compile any PB sources on my linux box I get the following error:

pbcompiler window.pb

******************************************
PureBasic Linux x86 v3.92
******************************************

Loading external modules...
Starting compilation...
40 lines processed ...

I wasn't questioning whether gcc was part of the compilation pipeline; I was calling your assumption that cygwin was necessary incorrect, which it is, and which any cursor examination of the documentation would reveal.

Not a defense of Max, just making sure the facts are straight.

You're 100% wrong, and you don't need to download cygwin to run BlitzMax on Windows platforms. Silly thing to say.

[Edit]While I'm at it, it's worthwhile to say that a a Bmax Sieve exe outperforms everything but MSVC on Windows -- including DevC -- and comes very close to equalling MSVC's ...

Ahhh now that's helped a bit. Thanks for helping to get the right syntax through my thick skull to Lars, freedimension and Al. Unfortunately I've still got problems using the routine as a DLL, but I'm not at all convinced that's Pure's fault, so it's time to do some more tedious line-by-line ...

Yep, I know -- and I can't do that. The allocation has to happen outside the procedure.

I can't modify the interface -- as I mentioned, this is the only way to interface with CallDLL, so the interface itself is non-negotiable.

Seriously, PokeB can't be the problem here, can it? I've double-checked the *out addresses after creation, after being passed to the procecure and just prior to ...

Ehrm... thanks for the reply, but that's definitely not the problem here unless I'm =really= missing something.

I'm at wit's end -- not saying much, I know, but still, it's as far as my wit can take me. The following code works if I've commented out the PokeB line near the end (clearly marked with the rem statement), but if I put the PokeB back in the whole thing siezes terribly -- ruining inner loop counts ...