This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: PowerPC floating point little-endian [7 of 15]
- From: Andreas Jaeger <aj at suse dot com>
- To: Alan Modra <amodra at gmail dot com>
- Cc: libc-alpha at sourceware dot org
- Date: Wed, 10 Jul 2013 12:52:53 +0200
- Subject: Re: PowerPC floating point little-endian [7 of 15]
- References: <20130710012435 dot GN2602 at bubble dot grove dot modra dot org> <20130710012828 dot GT2602 at bubble dot grove dot modra dot org> <51DCF7DF dot 5070303 at suse dot com> <20130710073221 dot GE2602 at bubble dot grove dot modra dot org>
On 07/10/2013 09:32 AM, Alan Modra wrote:
> On Wed, Jul 10, 2013 at 07:57:51AM +0200, Andreas Jaeger wrote:
>> On 07/10/2013 03:28 AM, Alan Modra wrote:
>>> More little-endian fixes.
>>>
>>> 2013-07-10 Anton Blanchard <anton@au1.ibm.com>
>>>
>>> * sysdeps/powerpc/bits/mathinline.h (__signbitf): Use builtin.
>>> (__signbit): Likewise. Correct for little-endian.
>>> (__signbitl): Call __signbit.
>>> (lrint): Correct for little-endian.
>>> (lrintf): Call lrint.
>>
>> Is calling lrint from lrintf really beneficial? Why not remove the
>> inline completely?
>
> We still want the inline, don't we? If you'd prefer the body of
> lrintf to duplicate lrint, I can do it that way..
Whether we need the inline is open for discussions, these are optional
optimizations,
Andreas
--
Andreas Jaeger aj@{suse.com,opensuse.org} Twitter/Identica: jaegerandi
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Jeff Hawn,Jennifer Guild,Felix Imendörffer,HRB16746 (AG Nürnberg)
GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126