This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: [RS6000] "G" and "H" constraints

From: Alan Modra <amodra at gmail dot com>
To: Segher Boessenkool <segher at kernel dot crashing dot org>
Cc: gcc-patches at gcc dot gnu dot org
Date: Thu, 29 Nov 2018 09:57:27 +1030
Subject: Re: [RS6000] "G" and "H" constraints
References: <20181125121400.GL21617@bubble.grove.modra.org> <20181128175520.GK23873@gate.crashing.org>

On Wed, Nov 28, 2018 at 11:55:21AM -0600, Segher Boessenkool wrote:
> >  (define_constraint "G"
> > -  "Constant that can be copied into GPR with two insns for DF/DI
> > -   and one for SF."
> > +  "Constant that can be copied into GPR with two insns for DF/DD
> > +   and one for SF/SD."
> >    (and (match_code "const_double")
> >         (match_test "num_insns_constant (op, mode)
> >  		    == (mode == SFmode ? 1 : 2)")))
> 
> The code does 2 for SDmode, the comment says 1.  Which is correct?

The comment is correct.  I'll fix the code to do
"mode == SFmode || mode == SDmode".

-- 
Alan Modra
Australia Development Lab, IBM

References:
- [RS6000] "G" and "H" constraints
  - From: Alan Modra
- Re: [RS6000] "G" and "H" constraints
  - From: Segher Boessenkool

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]