Re: [PATCH] include, opcodes: prefix REG_* macros in cris.h, nds32.h and microblaze-opcm.h

[Hans-Peter Nilsson <hp@bitrange.com>]

On Tue, 18 Aug 2026, Simon Marchi wrote:

> Bug 34537 reports:
> 
>     When building gdb with the default configure options of gdb buildbots (which
>     include --enable-targets=all), gdb fails to build:
> 
>       CXX    microblaze-tdep.o
>     In file included from ../../binutils-gdb/gdb/microblaze-tdep.c:37:

>       158 | #define REG_PC  REG_RIP
> 	  |         ^~~~~~
>     ../../binutils-gdb/gdb/../opcodes/microblaze-opcm.h:120:9: error: ?REG_SP? redefined [-Werror]
>       120 | #define REG_SP  1 /* stack pointer.  */
> 	  |         ^~~~~~

> Attempt to fix it by prefixing the macros with the arch name.

An alternative, less intrusive method, would be to #undef those 
macros in a suitable gdb toplevel file, maybe defs.h, after 
including cstdlib but before including binutils/gdb files.  The 
error is clearly on the Solaris side, for namespace intrusion 
and pollution.

Having said and suggested that, if that won't work, the CRIS 
bits are fine.
> 
> I only modified the macros starting with "REG_", but there are other
> macros with very generic names (e.g. "MAX_REG") that could be renamed
> too (for consistency, IMO, all the macros in those files should be
> prefixed the same way).
> 
> I couldn't build-test this patch on Solaris, but it build with
> --enable-targets=all on Linux.

How would you tell this attempt is complete?
(Maybe someone's doing evil ungreppable things like REG_ ## SP. ;-)
Are there not Solaris machines in the gcc build farm?

brgds, H-P