This is the mail archive of the gcc-help@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: Adding "soft-fp" support to a new target

From: Ian Lance Taylor <iant at google dot com>
To: Rathish C <Rathish dot C at kpitcummins dot com>
Cc: David Edelsohn <dje dot gcc at gmail dot com>, "gcc-help\ at gcc dot gnu dot org" <gcc-help at gcc dot gnu dot org>
Date: Wed, 19 May 2010 08:48:44 -0700
Subject: Re: Adding "soft-fp" support to a new target
References: <371569CBCFB2E745B891DBB88B2DFDDD19CA19996A@KCINPUNHJCMS01.kpit.com> <mcrd3wte1e7.fsf@dhcp-172-17-9-151.mtv.corp.google.com> <371569CBCFB2E745B891DBB88B2DFDDD19CA47753B@KCINPUNHJCMS01.kpit.com>

Rathish C <Rathish.C@kpitcummins.com> writes:

> I have one more doubt. Is there any way to obtain the following behavior?
> 	
> 	* By default the libgcc routines should be used.
> 	* If we specify the option say -msoft-fp during compilation, then the
>         soft-fp routines should be used instead of default libgcc routines.
>
> I have done the modifications to support the "-msoft-fp" option. 
> Please suggest how to proceed further.

Change the instruction predicates for your hard floating point
define_insn patterns so that they are not recognized if your option is
used.  E.g., referring back to rs6000, see the use of
TARGET_HARD_FLOAT and TARGET_SOFT_FLOAT in rs6000.md.

Ian

References:
- Adding "soft-fp" support to a new target
  - From: Rathish C
- Re: Adding "soft-fp" support to a new target
  - From: Ian Lance Taylor
- RE: Adding "soft-fp" support to a new target
  - From: Rathish C

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