Entertainingly enough, the C and C++ ABIs differ in this regard.<br>
<br>
C (even C99) says that sizeof(empty-struct) = 0, alignof(empty-struct) = 1<br>
C++ says that sizeof(empty-struct) = 1, alignof(empty-struct) = 1<br>
<br>
I say use the C layout, since Verrazano does its own layout for C++ that doesn't <br>
depend on what CFFI thinks the sizeof a type is.<br>
<br>
Sincerely, <br>
Rayiner hashem<br><br><div><span class="gmail_quote">On 9/7/05, <b class="gmail_sendername">Luis Oliveira</b> <<a href="mailto:luismbo@gmail.com">luismbo@gmail.com</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
On 7/set/2005, at 04:49, Kenny Tilton wrote:<br>> (cffi:defcstruct alcdevice-struct)<br>[...]<br>> I will let you all fight it out, but it looks as if CFFI needs to be<br>> more tolerant? I will try initializing the max-align to 2 and see how
<br>> much further I get.<br><br>CFFI needs bug fixing that's what it is. :-)<br><br>Well, I can't find anything specific about empty structures in the ABI<br>docs so I'm going to assume an empty struct indeed has size 0 and
<br>alignment 1 (or 0, whatever).<br><br>I'll add a test for this too.<br><br>--<br>Luís Oliveira<br><a href="http://student.dei.uc.pt/~lmoliv/">http://student.dei.uc.pt/~lmoliv/</a><br>Equipa Portuguesa do Translation Project
<br><a href="http://www2.iro.umontreal.ca/~pinard/po/registry.cgi?team=pt">http://www2.iro.umontreal.ca/~pinard/po/registry.cgi?team=pt</a><br><br>_______________________________________________<br>cffi-devel mailing list
<br><a href="mailto:cffi-devel@common-lisp.net">cffi-devel@common-lisp.net</a><br><a href="http://common-lisp.net/cgi-bin/mailman/listinfo/cffi-devel">http://common-lisp.net/cgi-bin/mailman/listinfo/cffi-devel</a><br></blockquote>
</div><br>