When visiting a file with non-latin unicode characters, most are
displayed wrong, e.g. the file includes a greek small lambda
(U+03BB), and emacs displays a small e with two dots U+00EB.
Some characters, e.g. U+03D1 (greek theta symbol), produce
a segmentation fault.
I use the "Bitstream vera sans mono" font.
Starting program: /usr/bin/emacs-snapshot-gtk -Q
[Thread debugging using libthread_db enabled]
[New Thread 47795254905600 (LWP 2647)]
[Switching to Thread 47795254905600 (LWP 2647)]
Breakpoint 3 at 0x498470: file xterm.c, line 8206.
Program received signal SIGSEGV, Segmentation fault.
x_get_font_repertory (f=<value optimized out>, font_info=<value optimized out>)
at xterm.c:10609