[09/12] ldbl-128ibm-compat: Provide nexttoward functions

Message ID 20180620020426.20372-10-tuliom@linux.ibm.com
State New
Headers show
Series
  • Introduce ieee128 symbols and redirections
Related show

Commit Message

Tulio Magno Quites Machado Filho June 20, 2018, 2:04 a.m.
From: "Gabriel F. T. Gomes" <gabriel@inconstante.eti.br>


The functions in the nexttoward family are special, in the sense that
they always have a long double argument, regardless of their suffix
(i.e.: nexttowardf and nexttoward have a long double argument, besides
the float and double arguments).

On top of that, they are also special because nexttoward functions are
not part of the _FloatN API, hence __nexttowardf128 do not exist.

This patch adds 4 new function implementations for the new long double
format:

  __nexttoward_to_ieee128
  __nexttowardf_to_ieee128
  __nexttowardieee128 (as an alias to __nextafterieee128)

2018-06-19  Gabriel F. T. Gomes  <gabriel@inconstante.eti.br>
	    Tulio Magno Quites Machado Filho  <tuliom@linux.ibm.com>

	* sysdeps/ieee754/ldbl-128ibm-compat/Versions: Add
	__nexttowardf_to_ieee128, __nexttowardieee128 and
	__nexttoward_to_ieee128.
	* sysdeps/ieee754/ldbl-128ibm-compat/s_nextafterf128.c: New file.
	* sysdeps/ieee754/ldbl-128ibm-compat/s_nexttowardf128.c: New file.
---
 sysdeps/ieee754/ldbl-128ibm-compat/Versions        |  3 ++
 .../ieee754/ldbl-128ibm-compat/s_nextafterf128.c   | 20 ++++++++++++
 .../ieee754/ldbl-128ibm-compat/s_nexttowardf128.c  | 38 ++++++++++++++++++++++
 3 files changed, 61 insertions(+)
 create mode 100644 sysdeps/ieee754/ldbl-128ibm-compat/s_nextafterf128.c
 create mode 100644 sysdeps/ieee754/ldbl-128ibm-compat/s_nexttowardf128.c

-- 
2.14.4

Comments

Joseph Myers June 20, 2018, 8:46 p.m. | #1
On Tue, 19 Jun 2018, Tulio Magno Quites Machado Filho wrote:

>   __nexttowardieee128 (as an alias to __nextafterieee128)


Is there an advantage to having this alias over having another #define in 
math.h like those used for drem and gamma?

-- 
Joseph S. Myers
joseph@codesourcery.com
Tulio Magno Quites Machado Filho June 21, 2018, 1:08 a.m. | #2
Joseph Myers <joseph@codesourcery.com> writes:

> On Tue, 19 Jun 2018, Tulio Magno Quites Machado Filho wrote:

>

>>   __nexttowardieee128 (as an alias to __nextafterieee128)

>

> Is there an advantage to having this alias over having another #define in 

> math.h like those used for drem and gamma?


No.  However, changing it breaks math/test-tgmath.c as it defines its own
nextafterl and nexttowardl in the same compilation unit, causing the assembler
to fail with:
/tmp/ccE9ypFK.s: Assembler messages:
/tmp/ccE9ypFK.s:22067: Error: symbol `__nextafterieee128' is already defined

Any suggestions on how to proceed?

-- 
Tulio Magno
Joseph Myers June 21, 2018, 4:25 p.m. | #3
On Wed, 20 Jun 2018, Tulio Magno Quites Machado Filho wrote:

> Joseph Myers <joseph@codesourcery.com> writes:

> 

> > On Tue, 19 Jun 2018, Tulio Magno Quites Machado Filho wrote:

> >

> >>   __nexttowardieee128 (as an alias to __nextafterieee128)

> >

> > Is there an advantage to having this alias over having another #define in 

> > math.h like those used for drem and gamma?

> 

> No.  However, changing it breaks math/test-tgmath.c as it defines its own

> nextafterl and nexttowardl in the same compilation unit, causing the assembler

> to fail with:

> /tmp/ccE9ypFK.s: Assembler messages:

> /tmp/ccE9ypFK.s:22067: Error: symbol `__nextafterieee128' is already defined

> 

> Any suggestions on how to proceed?


I suppose in that case adding the alias is reasonable.

-- 
Joseph S. Myers
joseph@codesourcery.com

Patch

diff --git a/sysdeps/ieee754/ldbl-128ibm-compat/Versions b/sysdeps/ieee754/ldbl-128ibm-compat/Versions
index 3c430f2074..01e67aec1a 100644
--- a/sysdeps/ieee754/ldbl-128ibm-compat/Versions
+++ b/sysdeps/ieee754/ldbl-128ibm-compat/Versions
@@ -83,6 +83,9 @@  libm {
     __nearbyintieee128;
     __nextafterieee128;
     __nextdownieee128;
+    __nexttowardf_to_ieee128;
+    __nexttowardieee128;
+    __nexttoward_to_ieee128;
     __nextupieee128;
     __powieee128;
     __remainderieee128;
diff --git a/sysdeps/ieee754/ldbl-128ibm-compat/s_nextafterf128.c b/sysdeps/ieee754/ldbl-128ibm-compat/s_nextafterf128.c
new file mode 100644
index 0000000000..419068f8b1
--- /dev/null
+++ b/sysdeps/ieee754/ldbl-128ibm-compat/s_nextafterf128.c
@@ -0,0 +1,20 @@ 
+/* Provide nexttowardl and nextafterl implementations for IEEE long double.
+   Copyright (C) 2018 Free Software Foundation, Inc.
+   This file is part of the GNU C Library.
+
+   The GNU C Library is free software; you can redistribute it and/or
+   modify it under the terms of the GNU Lesser General Public
+   License as published by the Free Software Foundation; either
+   version 2.1 of the License, or (at your option) any later version.
+
+   The GNU C Library is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+   Lesser General Public License for more details.
+
+   You should have received a copy of the GNU Lesser General Public
+   License along with the GNU C Library; if not, see
+   <http://www.gnu.org/licenses/>.  */
+#include "../float128/s_nextafterf128.c"
+
+strong_alias (__nextafterieee128, __nexttowardieee128)
diff --git a/sysdeps/ieee754/ldbl-128ibm-compat/s_nexttowardf128.c b/sysdeps/ieee754/ldbl-128ibm-compat/s_nexttowardf128.c
new file mode 100644
index 0000000000..4403783a56
--- /dev/null
+++ b/sysdeps/ieee754/ldbl-128ibm-compat/s_nexttowardf128.c
@@ -0,0 +1,38 @@ 
+/* Provide nexttoward[|f] implementations for IEEE long double.
+   Copyright (C) 2018 Free Software Foundation, Inc.
+   This file is part of the GNU C Library.
+
+   The GNU C Library is free software; you can redistribute it and/or
+   modify it under the terms of the GNU Lesser General Public
+   License as published by the Free Software Foundation; either
+   version 2.1 of the License, or (at your option) any later version.
+
+   The GNU C Library is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+   Lesser General Public License for more details.
+
+   You should have received a copy of the GNU Lesser General Public
+   License along with the GNU C Library; if not, see
+   <http://www.gnu.org/licenses/>.  */
+
+#include <float128_private.h>
+
+/* Build nexttoward functions with binary128 format.  */
+#undef weak_alias
+#define weak_alias(from, to)
+#undef libm_alias_ldouble
+#define libm_alias_ldouble(from, to)
+#undef __nexttoward
+#define __nexttoward __nexttoward_to_ieee128
+#include <sysdeps/ieee754/ldbl-128/s_nexttoward.c>
+
+#undef weak_alias
+#define weak_alias(from, to)
+#undef libm_alias_ldouble
+#define libm_alias_ldouble(from, to)
+#undef __nexttowardf
+#define __nexttowardf __nexttowardf_to_ieee128
+#include <sysdeps/ieee754/ldbl-128/s_nexttowardf.c>
+
+#include <libm-alias-ldouble.h>