Re: [patch] Resolve ppc64 func descriptors as .func (via .opd)

[Jan Kratochvil <jan dot kratochvil at redhat dot com>]

On Wed, 05 Dec 2012 17:30:35 +0100, Mark Wielaard wrote:
> Could you give an example of what you have in mind here? If you think it
> is relevant and/or might help me understand your use case.

I have tried to create such example now but I have found there is no way to
find out whether *SHNDXP belongs to mod->main or mod->debug.  Therefore the
parameter is useless.  And BTW elfutils really does not use it anywhere.


> OK, I think we can provide that through dwfl_module_addrsym () if we let
> the backend detect function descriptor symbols and return that
> symbol/name if the address matches. That is what I propose you
> implement.

This is already implemented in the first (second) post.


> > And I already tried to explain multiple times why the code symbols need to be
> > differently named and that such different name is already defined by ppc64 ABI.
> 
> Indeed, I still don't see why this is relevant. The address will match
> the function descriptor symbol and name. Not some synthetic code symbol.
> As far as I can see the returned name and symbol information is what the
> user wants/needs.

Maybe we still do not agree what the function should do.  I expected we have
already concluded dwfl_module_addrsym should return synthetic symbol for the
ppc64 instructions addresses and that we only disagree which name should be
reported for such returned synthetic symbol.

Sorry but this discussion is somehow dead-end, I do not understand what you
want to implement.  I can implement whatever you tell me but I have not found
yet a specification of implementable/working solution.


> > If we have not found an agreement on that part then the next step is to get an
> > approval from IBM.  I would find very unfriendly to break their ABI standard
> > in common standard package without even letting them know.
> 
> Please do ask if you think that is helpful. I might be missing why it is
> relevant/important. Having more opinions will certainly be welcome.

I expected that you asked for returning synthetic symbol only that the name
for the synthetic symbol should be dot-less.  In such case it is submittable
to IBM for approval.


> But I think just returning what is in the symbol table and not trying to
> munch the returned values from what we already have is good enough.

Should we mangle st_value to the first-function-instruction address?  In such
case it is not what is in symbol table and it is synthetic symbol required by
ppc64 ABI to be dot-named.  We can ask if IBM means their ABI spec seriously.


Thanks,
Jan