Re: [patch] Small fix for assigning values to vectors

["Ulrich Weigand" <uweigand@de.ibm.com>]

Ken Werner wrote:

> On Saturday, July 10, 2010 05:39:09 pm Daniel Jacobowitz wrote:
> > On Sat, Jul 10, 2010 at 03:03:53PM +0200, Ken Werner wrote:
> > > But while we are at it. Are there any objections on copying the contents
> > > to the destination instead of creating a pointer?
> > 
> > I think we're going a bit too far now, and maybe we need to figure out
> > what the semantics of internalvars are supposed to be...
> > 
> > With this, IIUC, "set $internalvar = program_array" is going to read
> > the whole array.  Previously, it would decay to a pointer as in C.
> 
> I agree. As I don't have a strong opinion either way on this particular case 
> I'd leave that decision to more experienced GDB developers.

Yes, this is a separate issue; let's leave the behaviour as is for now.
 
> Here is the patch I intended to post where only the call to 
> value_coerce_to_target is omitted and the value_must_coerce_to_target routine 
> returns zero in case of a vector. Tested on powerpc64-*-linux-gnu with no 
> regressions.

> Changelog:
> 
> 2010-07-09  Ken Werner  <ken.werner@de.ibm.com>
> 
> 	* valops.c (value_assign): Do not call to value_coerce_to_target.
> 	(value_must_coerce_to_target): Return 0 in case of TYPE_VECTOR.
> 
> testsuite/ChangeLog:
> 
> 2010-07-09  Ken Werner  <ken.werner@de.ibm.com>
> 
> 	* gdb.arch/altivec-abi.exp: New tests.

This is OK.

I think you should get Write After Approval access to the GDB CVS so that you
can commit your approved patches yourself.  Please apply using this form:
http://sourceware.org/cgi-bin/pdw/ps_form.cgi

Once you get access, please add yourself to the MAINTAINERS file under
the Write After Approval section, and then commit your patch.

Bye,
Ulrich

-- 
  Dr. Ulrich Weigand
  GNU Toolchain for Linux on System z and Cell BE
  Ulrich.Weigand@de.ibm.com