This is the mail archive of the gcc@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: Duplicate constraint characters in mn10300.h

From: Kazu Hirata <kazu at cs dot umass dot edu>
To: schwab at suse dot de
Cc: aoliva at redhat dot com, gcc at gcc dot gnu dot org
Date: Thu, 04 Sep 2003 10:51:27 -0400 (EDT)
Subject: Re: Duplicate constraint characters in mn10300.h
References: <20030904.082732.28806195.kazu@cs.umass.edu><je65k8wr19.fsf@sykes.suse.de>

Hi Andreas,

> > It looks like EXTRA_CONSTRAINT in mn10300.h has two 'S' there.  I am
> > afraid this is an oversight!?
> >
> > #define EXTRA_CONSTRAINT(OP, C) \
> >  ((C) == 'R' ? OK_FOR_R (OP) \
> >   : (C) == 'Q' ? OK_FOR_Q (OP) \
> >   : (C) == 'S' && flag_pic \
>                     ^^^^^^^^
> >   ? GET_CODE (OP) == UNSPEC && (XINT (OP, 1) == UNSPEC_PLT \
> > 				|| XINT (OP, 1) == UNSPEC_PIC) \
> >   : (C) == 'S' ? GET_CODE (OP) == SYMBOL_REF \
> >   : (C) == 'T' ? OK_FOR_T (OP) \
> >   : 0)
> 
> The fist one is for PIC, the other one is for non-PIC.  In the PIC case
> only UNSPEC is allowed (with a [SYMBOL_REF] as the first operand), the
> other case accepts unadorned SYMBOL_REFs.

Oops.  This is an oversight on my side then.  Thanks!

Kazu Hirata

References:
- Duplicate constraint characters in mn10300.h
  - From: Kazu Hirata
- Re: Duplicate constraint characters in mn10300.h
  - From: Andreas Schwab

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