Package: libffi5-dbg Version: 3.0.5-1 Severity: wishlist Trying to track down the current FTBFS in python-ctypes (turns out it's actually #480208 in libffi5, but I didn't think to look here first) and gdb can't trace through the assembly-source, eg. src/unix64.S. This can be fixed by setting AM_CCASFLAGS = $(AM_CPPFLAGS) -Wa,--gdwarf2 on line 171 of Makefile.am and rerunning automake. I checked, and dh_strip is successfully moving these symbols out to libffi5-dbg (tested by removing libfi5-dbg again and noting that gdb once again fails to resolve the assembly language file source) Feel free to reject this bug if this causes performance or other issues, I just thought it'd be a nice-to-have. (I'm guessing the reason gcc's -g doesn't already do this is for performance reasons? Does adding debugging symbols for the assembler code remove some optimisations?) -- System Information: Debian Release: lenny/sid APT prefers unstable APT policy: (500, 'unstable'), (1, 'experimental') Architecture: i386 (i686) Kernel: Linux 2.6.24-1-686 (SMP w/1 CPU core) Locale: LANG=en_AU.UTF-8, LC_CTYPE=en_AU.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash -- Paul "TBBle" Hampson, Paul.Hampson@Pobox.com Shorter .sig for a more eco-friendly paperless office.
Attachment:
pgpjg5nQvLiHN.pgp
Description: PGP signature