Bug#696556: ldd_load_address.diff
2013/7/20 Petr Salinger <Petr.Salinger@seznam.cz>:
>> 111_ldd_load_address.diff changes kernel-user ABI. How mergeable is this?
>
> It is not ABI chabge, it is only a change of
> implementation defined behaviour.
Seems like a moot point. The whole ABI is implementation-defined ;-)
> I do not know details of upstream toolchain.
> Does it support running binaries via dynamic linker, like
>
> /lib/ld-kfreebsd-x86-64.so.1 ./a.out
No. Their ldd just parses the ELF metadata.
The dynamic linker (/libexec/ld-elf.so.1) is not executable by itself.
> Compared to http://bugs.debian.org//cgi-bin/bugreport.cgi?bug=696556#15
> the final 111_ldd_load_address.diff only changes load address for DSO
> without program interpretter, i.e. mostly program interpretter (dynamic
> linker) itself.
So this change doesn't affect them? Then it shouldn't be hard to
persuade them to include it.
Otherwise we still have breakage for people running Debian as a chroot
under FreeBSD (or people using kfreebsd-downloader).
Will you send this patch to freebsd-arch? If not, I will do it. But it
is hard for me to defend a patch I don't fully understand.
Reply to: