* [PATCH] ldbl-128ibm: make ieee754.h work with IEEE 128 long double
@ 2020-02-11 17:12 Paul E. Murphy
2020-02-18 14:06 ` Tulio Magno Quites Machado Filho
0 siblings, 1 reply; 3+ messages in thread
From: Paul E. Murphy @ 2020-02-11 17:12 UTC (permalink / raw)
To: libc-alpha
Instead of attempting something more creative, just copy
the small struct from ldbl-128 and enable it when IEEE
long double is present, and update the ibm long double
variant if supported.
Likewise, provide a shadow copy of math_ldbl.h to prevent
the ibm128 specific long double header from poisoning
unrelated files due to it's usage in math_private.h.
---
.../ieee754/ldbl-128ibm-compat/math_ldbl.h | 30 ++++++++
sysdeps/ieee754/ldbl-128ibm/ieee754.h | 68 ++++++++++++++++++-
2 files changed, 96 insertions(+), 2 deletions(-)
create mode 100644 sysdeps/ieee754/ldbl-128ibm-compat/math_ldbl.h
diff --git a/sysdeps/ieee754/ldbl-128ibm-compat/math_ldbl.h b/sysdeps/ieee754/ldbl-128ibm-compat/math_ldbl.h
new file mode 100644
index 0000000000..352617e178
--- /dev/null
+++ b/sysdeps/ieee754/ldbl-128ibm-compat/math_ldbl.h
@@ -0,0 +1,30 @@
+/* Manipulation of the bit representation of 'long double' quantities.
+ Copyright (C) 2020 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
+ <https://www.gnu.org/licenses/>. */
+
+#ifndef _MATH_LDBL_H_IBM128_COMPAT
+#define _MATH_LDBL_H_IBM128_COMPAT 1
+
+#include <bits/floatn.h>
+
+/* Trampoline in the ldbl-128ibm headers if building against the
+ old abi. Otherwise, we have nothing to add. */
+#if __LONG_DOUBLE_USES_FLOAT128 == 0
+#include_next <math_ldbl.h>
+#endif
+
+#endif /* _MATH_LDBL_H_IBM128_COMPAT */
diff --git a/sysdeps/ieee754/ldbl-128ibm/ieee754.h b/sysdeps/ieee754/ldbl-128ibm/ieee754.h
index 87c64fc316..ef251613f5 100644
--- a/sysdeps/ieee754/ldbl-128ibm/ieee754.h
+++ b/sysdeps/ieee754/ldbl-128ibm/ieee754.h
@@ -21,6 +21,7 @@
#include <features.h>
#include <bits/endian.h>
+#include <bits/floatn.h>
__BEGIN_DECLS
@@ -111,6 +112,65 @@ union ieee754_double
#define IEEE754_DOUBLE_BIAS 0x3ff /* Added to exponent. */
+#if __LONG_DOUBLE_USES_FLOAT128 == 1
+/* long double is IEEE 128 bit */
+union ieee854_long_double
+ {
+ long double d;
+
+ /* This is the IEEE 854 quad-precision format. */
+ struct
+ {
+#if __BYTE_ORDER == __BIG_ENDIAN
+ unsigned int negative:1;
+ unsigned int exponent:15;
+ /* Together these comprise the mantissa. */
+ unsigned int mantissa0:16;
+ unsigned int mantissa1:32;
+ unsigned int mantissa2:32;
+ unsigned int mantissa3:32;
+#endif /* Big endian. */
+#if __BYTE_ORDER == __LITTLE_ENDIAN
+ /* Together these comprise the mantissa. */
+ unsigned int mantissa3:32;
+ unsigned int mantissa2:32;
+ unsigned int mantissa1:32;
+ unsigned int mantissa0:16;
+ unsigned int exponent:15;
+ unsigned int negative:1;
+#endif /* Little endian. */
+ } ieee;
+
+ /* This format makes it easier to see if a NaN is a signalling NaN. */
+ struct
+ {
+#if __BYTE_ORDER == __BIG_ENDIAN
+ unsigned int negative:1;
+ unsigned int exponent:15;
+ unsigned int quiet_nan:1;
+ /* Together these comprise the mantissa. */
+ unsigned int mantissa0:15;
+ unsigned int mantissa1:32;
+ unsigned int mantissa2:32;
+ unsigned int mantissa3:32;
+#else
+ /* Together these comprise the mantissa. */
+ unsigned int mantissa3:32;
+ unsigned int mantissa2:32;
+ unsigned int mantissa1:32;
+ unsigned int mantissa0:15;
+ unsigned int quiet_nan:1;
+ unsigned int exponent:15;
+ unsigned int negative:1;
+#endif
+ } ieee_nan;
+ };
+
+#define IEEE854_LONG_DOUBLE_BIAS 0x3fff /* Added to exponent. */
+#endif
+
+
+#if __LONG_DOUBLE_USES_FLOAT128 == 0 || __GNUC_PREREQ (7, 0)
/* IBM extended format for long double.
Each long double is made up of two IEEE doubles. The value of the
@@ -121,12 +181,16 @@ union ieee754_double
-0.0. No other requirements are made; so, for example, 1.0 may be
represented as (1.0, +0.0) or (1.0, -0.0), and the low part of a
NaN is don't-care. */
-
union ibm_extended_long_double
{
- long double ld;
+#if __LONG_DOUBLE_USES_FLOAT128 == 1 && __GNUC_PREREQ (7, 0)
+ __ibm128 ld;
+# else
+ long double ld;
+# endif
union ieee754_double d[2];
};
+#endif
__END_DECLS
--
2.21.1
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] ldbl-128ibm: make ieee754.h work with IEEE 128 long double
2020-02-11 17:12 [PATCH] ldbl-128ibm: make ieee754.h work with IEEE 128 long double Paul E. Murphy
@ 2020-02-18 14:06 ` Tulio Magno Quites Machado Filho
2020-02-21 14:32 ` Paul E Murphy
0 siblings, 1 reply; 3+ messages in thread
From: Tulio Magno Quites Machado Filho @ 2020-02-18 14:06 UTC (permalink / raw)
To: Paul E. Murphy, libc-alpha
"Paul E. Murphy" <murphyp@linux.vnet.ibm.com> writes:
> Instead of attempting something more creative, just copy
> the small struct from ldbl-128 and enable it when IEEE
> long double is present, and update the ibm long double
> variant if supported.
>
> Likewise, provide a shadow copy of math_ldbl.h to prevent
> the ibm128 specific long double header from poisoning
> unrelated files due to it's usage in math_private.h.
LGTM with a cosmetic change.
Reviewed-by: Tulio Magno Quites Machado Filho <tuliom@linux.ibm.com>
> @@ -121,12 +181,16 @@ union ieee754_double
> -0.0. No other requirements are made; so, for example, 1.0 may be
> represented as (1.0, +0.0) or (1.0, -0.0), and the low part of a
> NaN is don't-care. */
> -
> union ibm_extended_long_double
> {
> - long double ld;
> +#if __LONG_DOUBLE_USES_FLOAT128 == 1 && __GNUC_PREREQ (7, 0)
This is missing an indentation, like the else and endif clauses below.
--
Tulio Magno
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] ldbl-128ibm: make ieee754.h work with IEEE 128 long double
2020-02-18 14:06 ` Tulio Magno Quites Machado Filho
@ 2020-02-21 14:32 ` Paul E Murphy
0 siblings, 0 replies; 3+ messages in thread
From: Paul E Murphy @ 2020-02-21 14:32 UTC (permalink / raw)
To: Tulio Magno Quites Machado Filho, Paul E. Murphy, libc-alpha
Pushed with cosmetic fix after verification against all
build-many-glibcs.py targets. Thank you for your review Tulio.
On 2/18/20 8:05 AM, Tulio Magno Quites Machado Filho wrote:
> "Paul E. Murphy" <murphyp@linux.vnet.ibm.com> writes:
>
>> Instead of attempting something more creative, just copy
>> the small struct from ldbl-128 and enable it when IEEE
>> long double is present, and update the ibm long double
>> variant if supported.
>>
>> Likewise, provide a shadow copy of math_ldbl.h to prevent
>> the ibm128 specific long double header from poisoning
>> unrelated files due to it's usage in math_private.h.
>
> LGTM with a cosmetic change.
>
> Reviewed-by: Tulio Magno Quites Machado Filho <tuliom@linux.ibm.com>
>
>> @@ -121,12 +181,16 @@ union ieee754_double
>> -0.0. No other requirements are made; so, for example, 1.0 may be
>> represented as (1.0, +0.0) or (1.0, -0.0), and the low part of a
>> NaN is don't-care. */
>> -
>> union ibm_extended_long_double
>> {
>> - long double ld;
>> +#if __LONG_DOUBLE_USES_FLOAT128 == 1 && __GNUC_PREREQ (7, 0)
>
> This is missing an indentation, like the else and endif clauses below.
>
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2020-02-21 14:32 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-02-11 17:12 [PATCH] ldbl-128ibm: make ieee754.h work with IEEE 128 long double Paul E. Murphy
2020-02-18 14:06 ` Tulio Magno Quites Machado Filho
2020-02-21 14:32 ` Paul E Murphy
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).