Re: [linux-audio-user] Re: [linux-audio-dev] music engine

From: <torbenh@email-addr-hidden>
Date: Fri Apr 07 2006 - 20:44:30 EEST

On Fri, Apr 07, 2006 at 12:24:24PM +0200, David Olofson wrote:
> On Thursday 06 April 2006 23:59, Lee Revell wrote:
> > On Thu, 2006-04-06 at 21:45 +0000, carmen wrote:
> > >
> > > SC and Chuck: would likely work as well, but AFAIK neither one
> > > supports LADSPA or DSSI at the moment, SCLang definitely doesnt
> > > work
> > > on x86_64, and Chuck segfaulted as well but i'm guessing since its
> > > newer it was designed with 64bit compatibility and mind and the
> > > crash is some other issue
> > >
> >
> > Apps really don't have to be designed with 64 bit compatibility in
> > mind
> > - it should be rather simple to port them. Rather than ruling out
> > anything that doesn't work on AMD64 right now, I would advise you to
> > first select a set of apps that work in 32 bit mode, then find out
> > what it would take to port them.
> >
> > You might find that the CVS version supports AMD64, or that the
> > developer just has not seen a demand yet (the vast majority of
> > systems
> > are still 32 bit). Or maybe the developer still has a 32 bit
> > machine and has just never tried it.
>
> Indeed. I've been working on native AMD64 (Gentoo) for almost a year,
> and even though I spent most of my time hacking code for embedded 32
> bit systems, I compile and test all code as native 64 bit as well.
>
> The only real problem I can remember having was with the heap
> relocation in EEL, where some messy pointer arithmetics is needed.
> Other than that, I'm using the exact same source code for 32 bit and
> 64 bit targets; no special cases or anything.

>
> Basically, just don't assume that a pointer is 32 bits, and you're

but those bytecode interpreters of pd and such just assume, that int and
pointer are the same size...

well... perhaps it would work ok, if the integer was changed to int64
also...

dont ask me. i did not look at the code...

> fine. (That is, unless your programming style includes various
> inherently non-portable tricks. :-) Even in "naturally aligned"
> structs, you can usually get away with placing the pointers before
> any potentially smaller fields.
>
>
> //David Olofson - Programmer, Composer, Open Source Advocate
>
> .------- http://olofson.net - Games, SDL examples -------.
> | http://zeespace.net - 2.5D rendering engine |
> | http://audiality.org - Music/audio engine |
> | http://eel.olofson.net - Real time scripting |
> '-- http://www.reologica.se - Rheology instrumentation --'
>

-- 
torben Hohn
http://galan.sourceforge.net -- The graphical Audio language
Received on Sat Apr 8 00:15:02 2006

This archive was generated by hypermail 2.1.8 : Sat Apr 08 2006 - 00:15:02 EEST