Re: [PATCH] Support for Xilinx MicroBlaze

[Michael Eager <eager at eagercon dot com>]

Tom Tromey wrote:
> > > > > > "Michael" == Michael Eager <eager@eagercon.com> writes:
>
> Michael> When (if) this is fixed in gcc there will need to coordinate
> Michael> this with a corresponding fix in gdb.
>
> How would this be done?
> Users might be using an old gcc or a new one.

Not easy.  It would require coordination with users.  Lots.
Perhaps a test to see if code was generated by a particular version
of gcc or something equally ugly.

Even if gcc was fixed to accept reg+8 for the return address,
there would need to be a change to the MicroBlaze-specific
INCOMING_RETURN_ADDR_RTX for it to take any effect.  (There's
a warning note about this in gcc, which I might make more emphatic.)
So this is not going to happen by accident.

I'm just as happy to leave this problem unfixed.  Gcc generates an
incorrect CIE entry, which is philosophically not a good thing.
But fixing this causes far more problems than it solves.

> Michael> diff -urNp --exclude '*.swp' --exclude DEV-PHASE --exclude .svn --exclude CVS gdb/gdb/NEWS mb-gdb/gdb/NEWS
> Michael> --- gdb/gdb/NEWS	2009-10-14 11:15:51.000000000 -0700
> Michael> +++ mb-gdb/gdb/NEWS	2009-10-14 19:15:57.000000000 -0700
> Michael> @@ -484,6 +484,11 @@ Lattice Mico32                  lm32-*
> Michael>  x86 DICOS			i[34567]86-*-dicos*
> Michael>  x86_64 DICOS		        x86_64-*-dicos*
> Michael>  S+core 3			score-*-*
> Michael> +Xilinx MicroBlaze		microblaze-*-*
> Michael> +
> Michael> +* New Simulators
> Michael> +
> Michael> +Xilinx MicroBlaze		microblaze
>  
> I think this changes the "7.0" section, not the "since 7.0" section.

Oops.  The "Changes in GDB 7.0" was added after I updated the file
and I didn't notice it.  I'll move the entry.


--
Michael Eager	 eager@eagercon.com
1960 Park Blvd., Palo Alto, CA 94306  650-325-8077