Re: [Qemu-devel] [PATCH v5 10/10] softfloat: Use [u]int_fast64_t consistently

[Aurelien Jarno]

On Tue, Mar 08, 2011 at 12:10:50AM +0100, Andreas Färber wrote:
> Am 07.03.2011 um 10:56 schrieb Aurelien Jarno:
> 
> >On Mon, Mar 07, 2011 at 01:34:13AM +0100, Andreas Färber wrote:
> >>v5:
> >>* Initial.
> >>
> >>Signed-off-by: Andreas Färber <address@hidden>
> >>---
> >>fpu/softfloat-native.c |   26 +++++++++++++-------------
> >>fpu/softfloat-native.h |   28 ++++++++++++++--------------
> >>fpu/softfloat.c        |    4 ++--
> >>3 files changed, 29 insertions(+), 29 deletions(-)
> 
> >>diff --git a/fpu/softfloat-native.h b/fpu/softfloat-native.h
> >>index 6cf5dc3..b198f48 100644
> >>--- a/fpu/softfloat-native.h
> >>+++ b/fpu/softfloat-native.h
> 
> >>@@ -392,8 +392,8 @@ INLINE float64 float64_scalbn(float64 a, int n)
> >>*----------------------------------------------------------------------------*/
> >>int_fast32_t floatx80_to_int32( floatx80 STATUS_PARAM );
> >>int_fast32_t floatx80_to_int32_round_to_zero( floatx80
> >>STATUS_PARAM );
> >>-int64_t floatx80_to_int64( floatx80 STATUS_PARAM);
> >>-int64_t floatx80_to_int64_round_to_zero( floatx80 STATUS_PARAM);
> >>+int64_fast_t floatx80_to_int64( floatx80 STATUS_PARAM);
> >>+int64_fast_t floatx80_to_int64_round_to_zero( floatx80
> >>STATUS_PARAM);
> >
> >This type doesn't exist, so the code doesn't even compile...
> 
> I did compile-tested these, so apparently my host or targets do not
> use softfloat-native... It was late so I can't rule out other
> mistakes there.

softfloat-native is used on i386 and x86_64 targets, which are enabled
by default when running ./configure.

> Would you prefer to treat softfloat-native as a separate API and not
> touch it in this series?
> 

In my opinion, if we do such change, we should do them in both 
softfloat and softfloat-native.

-- 
Aurelien Jarno                          GPG: 1024D/F1BCDB73
address@hidden                 http://www.aurel32.net