This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH v2 8/8] float128: Add wrappers to override ldbl-128 as float128.
- From: Joseph Myers <joseph at codesourcery dot com>
- To: "Gabriel F. T. Gomes" <gftg at linux dot vnet dot ibm dot com>
- Cc: <libc-alpha at sourceware dot org>
- Date: Tue, 2 May 2017 21:51:07 +0000
- Subject: Re: [PATCH v2 8/8] float128: Add wrappers to override ldbl-128 as float128.
- Authentication-results: sourceware.org; auth=none
- References: <1493415280-30534-1-git-send-email-gftg@linux.vnet.ibm.com> <1493415280-30534-9-git-send-email-gftg@linux.vnet.ibm.com>
On Fri, 28 Apr 2017, Gabriel F. T. Gomes wrote:
> +/* public entry points. */
> +#define asinhl asinhf128
> +#define atanl atanf128
> +#define atanl atanf128
> +#define cbrtl cbrtf128
> +#define ceill ceilf128
> +#define ceill ceilf128
> +#define copysignl copysignf128
> +#define copysignl copysignf128
> +#define cosl cosf128
> +#define cosl cosf128
You have duplicate definitions of several macros (atanl ceill copysignl
cosl).
> +#define finitel finitef128
> +#define isinfl isinff128
> +#define isnanl isnanf128
These only have public names for long double, not for _Float128; it might
be better to rename to *do_not_use names as for nexttoward. (In the
nexttoward case the function shouldn't be built at all for _Float128;
likewise scalb, significand. Of course you might later need to build them
for long double = float128, but public *f128 interfaces still wouldn't be
exported for them. In the cases of finite, isinf, isnan, the __*f128
interfaces exist, but no public ABI without the leading __.)
> +#define fpclassifyl fpclassifyf128
> +#define issignalingl issignalingf128
And fpclassifyl, issignalingl don't exist; no need for these #defines.
--
Joseph S. Myers
joseph@codesourcery.com