This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH] BZ#887 logb[l](1) shows negative sign on zero inFE_DOWNWARD rounding mode
- From: "Ryan S. Arnold" <ryan dot arnold at gmail dot com>
- To: Andreas Jaeger <aj at suse dot com>
- Cc: libc-alpha at sourceware dot org
- Date: Fri, 27 Apr 2012 11:28:30 -0500
- Subject: Re: [PATCH] BZ#887 logb[l](1) shows negative sign on zero inFE_DOWNWARD rounding mode
- References: <1335467805.5328.87.camel@localhost.localdomain><4F9A4540.9000800@suse.com>
On Fri, Apr 27, 2012 at 2:05 AM, Andreas Jaeger <aj@suse.com> wrote:
>> + Âif (!fesetround (FE_DOWNWARD))
>> + Â Â{
>> +
>> + Â Â Â/* IEEE 754-2008 says (section 5.3.3) that "logB(1) is +0.". ÂLibm
>> + Â Â Â * should not return -0 from logb in any rounding mode. ÂPowerPC32
>> has
>> + Â Â Â * failed with this test for power4 logb (and logbl on all PowerPC
>> + Â Â Â * platforms) in the past due to instruction selection. ÂGCC PR
>> 52775
>> + Â Â Â * provides the availability of the fcfid insn in 32-bit mode which
>> + Â Â Â * eliminates the use of fsub in this instance and prevents the
>> negative
>> + Â Â Â * signed 0.0. Â*/
>
>
> Please remove the leading '*' and reformat.
>
>
>> + Â Â Â/* BZ #887 .*/
>
>
> Two spaces after the "."
>
> Ok with those changes.
I've made the requested changes and pushed upstream with the following commit:
commit a462cb63326595eb018e26ea415f8ddc1ed50730
Ryan S. Arnold