This is the mail archive of the
libc-alpha@sources.redhat.com
mailing list for the glibc project.
Re: ppc64 vDSO in mainline
- From: Steve Munroe <sjmunroe at us dot ibm dot com>
- To: Alan Modra <amodra at bigpond dot net dot au>
- Cc: Benjamin Herrenschmidt <benh at kernel dot crashing dot org>, libc-alpha at sources dot redhat dot com
- Date: Tue, 29 Mar 2005 09:00:18 -0600
- Subject: Re: ppc64 vDSO in mainline
Alan Modra <amodra@bigpond.net.au> wrote on 03/28/2005 06:46:12 PM:
> On Sat, Mar 26, 2005 at 10:13:58AM +1100, Benjamin Herrenschmidt wrote:
> [About function pointers]
> > But ppc64 will always load pointers from OPDs no ?
>
> Yes. This is why glibc/VDSO must provide relocated OPD entries if
> VDSO function symbols are to directly resolve references in a user app,
> ie. so that calls can be made without bouncing through some glibc stub
> code. Calls via a function pointer don't go through the PLT.
>
But inter-library bindings do (use the PLT) and this is what I am trying
to enable. See powerpc64/dl-machine.h:elf_machine_fixup_plt to how we
handle unrelocated .opd when resolving a PLT entry.
Steven J. Munroe
Linux on Power Toolchain Architect
IBM Corporation, Linux Technology Center