Re: PATCH for [ARM] subsequent use of plus and minus operators could be improved

[Richard Earnshaw <rearnsha at arm dot com>]

> Richard Earnshaw wrote:
>  >>Should I attach the patch for the generic solution of the
>  >>problem?
>  >
>  >
>  > We need to see the patch.
>  >
>  > R.
> 
> The patch is attached.
> This algorithm works in the postreload phase (register
> allocation algorithms produce the unnecessary plus/minus
> insn which is described in PR 10242). If I misplaced the
> code, correct me, please!
> 
> I've only one comment on using this algorithm. It doesn't
> test if the combined constant can be stored in the target
> specific instruction. The flow2 phase is responsible for
> this testing, and it corrects the problem if necessary.


First of all, please read the documents on contributing to GCC and pay 
particular attention to the coding standards sections.  Patches can only 
be accepted if they conform to these standards.

	http://gcc.gnu.org/contribute.html

You should look to see if this code could be merged with one of the other 
reload_cse passes so that we can avoid another scan through the entire RTL 
of a function.  One possible candidate here is reload_cse_move2add which 
is looking for similar types of thing to fix up.

validate_change() can tell you if the substitution you made is correct 
(and not do the transformation if the insn is not valid).  You should only 
kill the old insn if the change is successful.

> 
>  >
>  > BTW.  Have you filed a copyright assignment with the FSF?
>  >
> 
> We have sent the request.

You need to contact the Copyright Assignment Clerk at the FSF (IIRC 
assign at gnu dot org).  Explain what you want to contribute to and who will be 
contributing; they can then send you the correct papers for you to sign.

R.