This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
RE: SH relocation differences in older versions of the binutils
- From: Jeff Baker <jbaker at qnx dot com>
- To: Jeff Baker <jbaker at qnx dot com>
- Cc: "'binutils at sources dot redhat dot com'" <binutils at sources dot redhat dot com>
- Date: Wed, 17 Dec 2003 13:41:45 -0500
- Subject: RE: SH relocation differences in older versions of the binutils
Can someone help me figure out some code that I can use to have the GNU
tools produce a binary with the following relocations?
R_SH_SWITCH8
R_SH_GNU_VTINHERIT
R_SH_GNU_VTENTRY
R_SH_LOOP_START
R_SH_LOOP_END
I need an object to use as a testcase for our efforts.
> -----Original Message-----
> From: Jeff Baker
> Sent: Tuesday, December 16, 2003 3:26 PM
> To: Jeff Baker
> Cc: 'binutils@sources.redhat.com'
> Subject: RE: SH relocation differences in older versions of the binutils
>
> Turns out that this isn't exactly the problem we're having. We're still
> using 2.12.1, which is well before these changes were made. Also, as a
> happy coincidence, none of the relocs that we emit seem to have moved so
> hopefully we can gracefully begin to use a new version of the binutils in
> the near future without this causing us headaches.
>
> The problem we're having boils down to what seems to be a fix for the way
> that the SH ld handled addends in 2.10.1. Specifically the changes
> proposed in this posting: http://sources.redhat.com/ml/binutils/2001-
> 09/msg00302.html
>
> To help us work with both of these issues we're wondering a couple of
> things.
>
> 1) Is there currently, or should there be, some sort of indicator
> included in object files that specifies which version of the binutils it
> was produced with?
>
> 2) Is there a reliable method of determining whether an SH object file is
> using the addends correctly? We want to be able to detect and warn, or
> potentially correct older objects.
>
> > A few questions, if I may.
> >
> > Would I be correct in saying that the reason that the ABIVERSION was not
> > incremented with this change is because it was ultimately considered a
> > huge bug fix instead of a new ABI?
> >
> > Aside from parsing the relocs and attempting to determine which version
> of
> > the table they're using, can anyone recommend a way to determine whether
> > an object uses the old relocs or new? Right now we're looking for a way
> > to dynamically translate the reloc table to the new relocs at link time.
> >
> > > AHA. That's exactly what I was looking for, thanks.
> > >
> > > > -----Original Message-----
> > > > From: DJ Delorie [mailto:dj@redhat.com]
> > > > Sent: Friday, December 12, 2003 2:30 PM
> > > > To: jbaker@qnx.com
> > > > Cc: binutils@sources.redhat.com
> > > > Subject: Re: SH relocation differences in older versions of the
> > binutils
> > > >
> > > >
> > > > > Does anyone know/remember what would have changed with the
> handling
> > of
> > > > > relocations for SH between these two versions of the binutils?
> > > >
> > > > You'll want to read through this thread:
> > > >
> > > > http://sources.redhat.com/ml/binutils/2003-08/msg00271.html