Hi Uli,

Ulrich Weigand schrieb:
> 
> Finally, I do think the change in inself is correct (provided the comments
> are adapted).  In fact, the very same should be done to the other two
> functions in gdbarch.sh that use the "local current_gdbarch variable" hack,
> verify_gdbarch and gdbarch_dump.

thank you very much. Now I understand what was meant. I reworked the patch to update
the comments and also rework verify_gdbarch and gdbarch_dump.

Tested on x86 without regression. 

Would this be ok to commit ?

ChangeLog:

	* gdbarch.sh (gdbarch_alloc): Replace current_gdbarch by gdbarch. Remove
	obsolete comment.
	(verify_gdbarch, gdbarch_dump): Likewise.
	(do_read): Update comment.
	(addr_bit, bfd_arch_info, target_desc, long_long_bit, floatformat)
	(ptr_bit, name_of_malloc): Replace current_gdbarch by gdbarch.
	* gdbarch.{c,h}: Regenerate.


-- 
Markus Deuling
GNU Toolchain for Linux on Cell BE
deuling@de.ibm.com