This is the mail archive of the
libc-alpha@sources.redhat.com
mailing list for the glibc project.
Re: ppc relocs
- From: Roland McGrath <roland at redhat dot com>
- To: Alan Modra <amodra at bigpond dot net dot au>
- Cc: libc-alpha at sources dot redhat dot com, Paul Mackerras <paulus at samba dot org>,Steven Munroe <sjmunroe at us dot ibm dot com>
- Date: Sun, 2 Mar 2003 16:16:00 -0800
- Subject: Re: ppc relocs
> ppc32 does. gcc currently always produces PIC code for ppc64.
My question is about the ABI spec, not just current practice. It sounds
like it is in fact kosher on ppc32 to have all those 16-bit relocs in a
dynamic reloc section and ld.so should handle it. Is that also true on ppc64?
> True enough I suppose for ppc64. I know Steve has some cleanups for
> ppc64 dl-machine.h in the pipeline, eg. remove all DTPREL16 reloc
> support.
Well, Steve should be sending his patches more often! I've checked in the
(untested) cleanup I did because it was so crufty as it was. Now Steve
will have to merge his version with that.
> What you're missing is the description of the half16ds field, which
> is a half16 field without the bottom two bits. So we calculated the
> field value with ">> 2" then install to a 16 bit word with "<< 2", or
> equivalently, forget about the shifts and instead mask.
I see. Thanks for the explanation. I figured it must be something like
this, but couldn't quite find it clearly said in the spec.
Thanks,
Roland