This is the mail archive of the
libffi-discuss@sourceware.org
mailing list for the libffi project.
Re: [patch] fix libffi build on arm-linux-gnueabi with -mfloat-abi=softfp
- From: Andrew Haley <aph at redhat dot com>
- To: Matthias Klose <doko at ubuntu dot com>
- Cc: GCJ-patches <java-patches at gcc dot gnu dot org>, GCC Patches <gcc-patches at gcc dot gnu dot org>, libffi-discuss at sourceware dot org
- Date: Wed, 23 Sep 2009 15:44:30 +0100
- Subject: Re: [patch] fix libffi build on arm-linux-gnueabi with -mfloat-abi=softfp
- References: <4ABA33BF.8090205@ubuntu.com>
Matthias Klose wrote:
> Based on the comments in PR41443:
>
> All the references to __SOFTFP__ in this file are incorrect for EABI; the
> __SOFTFP__ code should be used for *soft-float ABI* whether or not VFP is
> enabled, and __SOFTFP__ does specifically mean soft-float not soft-float
> ABI.
>
> If you change the conditionals to __SOFTFP__ || __ARM_EABI__ then
> -mfloat-abi=softfp should work. -mfloat-abi=hard won't; that would need
> both a new macro to identify the hard-VFP ABI, and new libffi code to
> handle that ABI.
>
> No regressions when running the libffi testsuite with and without
> -mfloat-abi=softfp.
>
> Ok for the trunk, ok for the branches (checking that the macros are
> available there)?
Yes, I think that's the right thing to do.
Andrew.