Re: [PATCH]: Fixes for pseudo regs support

[Kevin Buettner <kevinb at cygnus dot com>]

On Nov 21, 11:55pm, Stephane Carrez wrote:

> Andrew Cagney wrote:
> > 
> > A pseudo register can't lie in memory as it is constructed from one or
> > more real registers.  The frame info code needs to be able to construct,
> > on the fly, a pseudo register from a combination of both saved registers
> > and real registers.
> 
> Ok. So, I'm back to my initial problem now.

My apologies for coming to this discussion late, but I had this
problem too on the IA-64 in which some of the registers (the so
called stacked registers) are only obtainable from memory.  

The way that I approached it (which predates the pseudo register
support) is to define ia64_get_saved_register() to do, umm,
interesting things.  (There were other registers too which
required special attention.)

I would like to convert this code to use pseudo registers though some
day.  FWIW, I think it should be possible to define pseudo registers
which interogate memory to get/set their values.

Kevin