On Oct 27, 2005, at 6:33 PM, Manfred Bergmann wrote:
These new releases bring experimental Intel compatibility to both the
CamelBones framework and ShuX. In addition, there have been a few
minor bug fixes and additions to CamelBones - see the included
release notes for details.
I've read in earlier posts that you were concerned about the intel
port because of something with libff (libffi).
It's ffcall actually - ffi is another library that does the same thing.
<http://www.haible.de/bruno/packages-ffcall.html>
I use the avcall() function to build up a stack frame with which to
call the C function objc_msgSend(). Because it relies on low-level
ABI details like stack and argument alignment, it's more troublesome
to port than a typical Cocoa or Carbon app that uses high-level
library calls.
Does this version mean, you have have a solution to this and fixed
or implemented it already?
It's a potential solution. I don't have an Intel Mac to test whether
it runs or not, but I've got the build procedure pretty much ironed
out. And, I have a good idea about what the potential problem areas
might be.
Note that one problem has already been reported and fixed. There's a
newer ShuX snapshot on the web site:
<http://camelbones.sourceforge.net/download/cvs-install.html>
sherm--
Cocoa programming in Perl: http://camelbones.sourceforge.net
Hire me! My resume: http://www.dot-app.org
|
