Re: unicode-2 branch segfaulting on MacOS X?

[Geoffrey Alan Washburn]

Kenichi Handa wrote:
> In article <address@hidden>, Geoffrey Alan Washburn <address@hidden> writes:
>
> > I've been trying to get the latest revision of the emacs-unicode-2 
> > branch to build and run under MacOS X.  Building seems to work fine, 
> > I've been using the following configuration
>
> > ./configure --prefix=/local --enable-font-backend --with-gtk --with-xft 
> > --with-tiff --with-ungif --with-xpm --enable-asserts CPPFLAGS=-idirafter 
> > /usr/X11R6/include
>
> > However, when I attempt to invoke emacs with
>
> > emacs --enable-font-backend --font "Bitstream Vera Sans Mono-16"
>
> > it seems to be segfaulting somewhere in font_unparse_fcname.  Any ideas?
>
> I can't reproduce it on GNU/Linux.  

	Indeed, the same version works fine on my ThinkPad running Debian 
GNU/Linux.

> So, could you please use gdb to find exactly where (and how) Emacs causes 
> segfault
> (see etc/DEBUG if you are not familiar with gdb).

	Could you be more specific about what you mean by "how" it is 
segfaulting?  It is pretty obvious that it is reading/writing an invalid 
memory location.  As I said the problem occurs in font_unparse_fcname. 
In particular, line 1258 of font.c with parameters font=42759700, 
pixel_size=19, name=0x2a20d20 "", and nbytes=96.