[Toybox] ToyBox ?: gcc extension excision

[Rob Landley]

On 04/09/2012 09:46 AM, Kevin Chase wrote:
> I was looking into using toysh on Windows to solve a porting issue and
> this cropped up with the MS compiler.

I note that toysh needs a _lot_ of work still.  I have extensive design
notes for pending todo items, but haven't coded most of it up yet.  (I
don't think it even has environment variable support yet.)

> I wouldn't be surprised if this is an issue with Unix vendor compilers
> as well.

Are there any that still matter? Back in the 90's the "86open" project
tried to come up with a standardized binary format for x86 unix (first
meeting August 22, 1997), and after SCO developer Michael Davidson wrote
the "lxrun" utility to run Linux binaries on Unixware, and then Sun
ported it to Solaris, and the 86open project concluded (final report
dated July 25, 1999) saying that Linux binaries were the de-facto
portable x86 binary standard.  Since then BSD developed a linux binary
layer, AIX came out with AIX 5L (with the L standing for Linux binary
compatability), and so on...

This was long before Sun's corpse was bought by Oratroll, SGI
end-of-lifed Irix in favor of Linux, SCO got bought by a Linux vendor
(Caldera) and then exploded into a flurry of patenty litigation (just
like drowning swimmers drag others down with them as they flail around)...

Could you be a little more specific about _which_ vendor?  The Open64
compiler got open sourced (I think that was Dec Alpha, back before Tru64
died)...

(I'm honestly curious, I was researching this for Halloween IX back when
SCO jr. was suing the world at large for not supplying it with a viable
ongoing business model that insulated it from its own stupidity.)

*shrug*  The patch is merged either way...

Rob
-- 
GNU/Linux isn't: Linux=GPLv2, GNU=GPLv3+, they can't share code.
Either it's "mere aggregation", or a license violation.  Pick one.