Re: [Gcl-devel] 2.6.2 benchmarks

[Camm Maguire]

Greetings!

"Mike Thomas" <address@hidden> writes:

> Hi Camm.
> 
> Thanks for that.  I was worried you were going to say an 800 MHz PIII!
> 
> | Anyone know why clisp is not compiling the browse test (and corman the
> | others?)
> 
> corman - from memory FFT-MOD died because of a syntax incompatibility for
> doubles;  STAK just hung for half an hour twice over so I gave up on it.
> 
> I saw yesterday that ECL has made it onto Windows via MinGW32 so it will be
> good to try it out at some stage too.  It uses DLLs for dynamic loading
> rather than object linking, so I suppose that is roughly equivalent with our
> Unix dlopen method but with fully sealed executable modules (which is what
> DLL's are on Windows).  That was the approach used in the Windows CE GCL
> port too.
> 

Great!  This probably means that a GCL dlopen build is not too far
away.  Good to know in case we expect future difficulties with the
linker as a cross-check.  However, the native object relocation we
have currently working on mingw is superior in a number of respects,
so at least for this we should be thankful :-).

Take care,

> Cheers
> 
> Mike Thomas
> 
> 
> 
> 

-- 
Camm Maguire                                            address@hidden
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah