This is the mail archive of the
libc-alpha@sources.redhat.com
mailing list for the glibc project.
Re: [PATCH] PPC64 tls fixes
- From: Roland McGrath <roland at redhat dot com>
- To: Alan Modra <amodra at bigpond dot net dot au>
- Cc: sjmunroe at us dot ibm dot com, libc-alpha at sources dot redhat dot com
- Date: Tue, 4 Mar 2003 23:33:05 -0800
- Subject: Re: [PATCH] PPC64 tls fixes
> On Tue, Mar 04, 2003 at 07:37:50PM -0800, Roland McGrath wrote:
> > > However (at least for ppc64) the DTPREL16* relocs will never generate
> > > dynamic endties.
> >
> > I'm assuming this is a typo for "entries" and you meant dynamic relocs.
> > I asked about this earlier and Alan Modra answered my question on Monday
> > with this answer contrary to what you've just said:
> >
> > > As far as I'm aware, there's nothing in the ppc64 ABI that prohibits
> > > position dependent code. Thus it's valid to generate the ADDR16 relocs
> > > in a shared library.
>
> ADDR16 != DTPREL16
Well, the 16-bit TLS relocs were what I asked about, perhaps it wasn't
clear. Would those not also be generated in any non-PIC code doing
IE-model TLS access?