This is the mail archive of the
mailing list for the binutils project.
Re: SH FDPIC ABI spec/binutils and kernel conflict on flag definitions
- From: Rich Felker <dalias at libc dot org>
- To: David Howells <dhowells at redhat dot com>
- Cc: Nick Clifton <nickc at redhat dot com>, linux-sh at vger dot kernel dot org, binutils at sourceware dot org, Oleg Endo <oleg dot endo at t-online dot de>, Steven Miao <realmz6 at gmail dot com>
- Date: Thu, 10 Sep 2015 12:01:49 -0400
- Subject: Re: SH FDPIC ABI spec/binutils and kernel conflict on flag definitions
- Authentication-results: sourceware.org; auth=none
- References: <20150910155059 dot GO17773 at brightrain dot aerifal dot cx> <20150910145808 dot GN17773 at brightrain dot aerifal dot cx> <20150910033400 dot GM17773 at brightrain dot aerifal dot cx> <55F17015 dot 8090207 at redhat dot com> <23476 dot 1441899930 at warthog dot procyon dot org dot uk> <23551 dot 1441900415 at warthog dot procyon dot org dot uk>
On Thu, Sep 10, 2015 at 04:53:35PM +0100, David Howells wrote:
> Rich Felker <firstname.lastname@example.org> wrote:
> > On the other hand, the only existing way to produce a binary that both
> > (1) needs constant displacement, and (2) actually gets constant
> > displacement from the kernel at load time, is to manually edit the ELF
> > headers to flip the bit. So I really doubt any such binaries exist. Do
> > you have a reason to believe they do?
> Well, Fujitsu asked for it for FRV - I've no idea whether they have such
> binaries still.
OK. Can we get feedback from anyone who's actually using this code
now? I don't see how the current situation is at all acceptable for
anyone using it, but maybe I'm missing something. It seems programs
built as FDPIC are going through all the ABI cost for FDPIC (function
descriptors, restricted relative addressing) for absolutely no gain
because the kernel just loads them like plain non-FDPIC ELF files...