public inbox for libstdc++-cvs@sourceware.org
help / color / mirror / Atom feed
From: Jonathan Wakely <redi@gcc.gnu.org>
To: gcc-cvs@gcc.gnu.org, libstdc++-cvs@gcc.gnu.org
Subject: [gcc r12-10244] libstdc++: Fix std::abs(__float128) for -NaN and -0.0 [PR109758]
Date: Mon, 18 Mar 2024 14:05:21 +0000 (GMT) [thread overview]
Message-ID: <20240318140521.A97DA385840C@sourceware.org> (raw)
https://gcc.gnu.org/g:9f381ebb7211c1359f5de87760148096fcba3357
commit r12-10244-g9f381ebb7211c1359f5de87760148096fcba3357
Author: Jonathan Wakely <jwakely@redhat.com>
Date: Wed May 10 12:20:58 2023 +0100
libstdc++: Fix std::abs(__float128) for -NaN and -0.0 [PR109758]
The current implementation of this non-standard overload of std::abs
incorrectly returns a negative value for negative NaNs and negative
zero, because x < 0 is false in both cases.
Use fabsl(long double) or fabsf128(_Float128) if those do the right
thing. Otherwise, use __builtin_signbit(x) instead of x < 0 to detect
negative inputs. This assumes that __builtin_signbit handles __float128
correctly, but that seems to be true for all of GCC, clang and icc.
libstdc++-v3/ChangeLog:
PR libstdc++/109758
* include/bits/std_abs.h (abs(__float128)): Handle negative NaN
and negative zero correctly.
* testsuite/26_numerics/headers/cmath/109758.cc: New test.
(cherry picked from commit af595613acbd9863198ae69c7b1c9e856bca9e4f)
Diff:
---
libstdc++-v3/include/bits/std_abs.h | 13 +++++-
| 52 ++++++++++++++++++++++
2 files changed, 63 insertions(+), 2 deletions(-)
diff --git a/libstdc++-v3/include/bits/std_abs.h b/libstdc++-v3/include/bits/std_abs.h
index c8d589d2b0a..bd14addd159 100644
--- a/libstdc++-v3/include/bits/std_abs.h
+++ b/libstdc++-v3/include/bits/std_abs.h
@@ -101,11 +101,20 @@ _GLIBCXX_BEGIN_NAMESPACE_VERSION
__extension__ inline _GLIBCXX_CONSTEXPR
__float128
abs(__float128 __x)
- { return __x < 0 ? -__x : __x; }
+ {
+#if defined(_GLIBCXX_LDOUBLE_IS_IEEE_BINARY128)
+ return __builtin_fabsl(__x);
+#elif defined(_GLIBCXX_HAVE_FLOAT128_MATH)
+ return __builtin_fabsf128(__x);
+#else
+ // Assume that __builtin_signbit works for __float128.
+ return __builtin_signbit(__x) ? -__x : __x;
+#endif
+ }
#endif
_GLIBCXX_END_NAMESPACE_VERSION
} // namespace
-} // extern "C"++"
+} // extern "C++"
#endif // _GLIBCXX_BITS_STD_ABS_H
--git a/libstdc++-v3/testsuite/26_numerics/headers/cmath/109758.cc b/libstdc++-v3/testsuite/26_numerics/headers/cmath/109758.cc
new file mode 100644
index 00000000000..c9716d3d372
--- /dev/null
+++ b/libstdc++-v3/testsuite/26_numerics/headers/cmath/109758.cc
@@ -0,0 +1,52 @@
+// { dg-do run }
+// PR libstdc++/109758 std::abs(__float128) doesn't support NaN
+
+#include <cmath>
+#include <testsuite_hooks.h>
+
+#if !defined(__STRICT_ANSI__) && defined(_GLIBCXX_USE_FLOAT128)
+void
+test_nan()
+{
+ __float128 nan = __builtin_nanl("");
+ VERIFY( !__builtin_signbit(std::abs(nan)) );
+ VERIFY( !__builtin_signbit(std::abs(-nan)) );
+}
+
+void
+test_zero()
+{
+ __float128 zero = 0.0;
+ VERIFY( !__builtin_signbit(std::abs(zero)) );
+ VERIFY( !__builtin_signbit(std::abs(zero * -2.0)) );
+}
+
+void
+test_neg()
+{
+ VERIFY( std::abs((__float128)-1.0) == -1.0 );
+ VERIFY( std::abs((__float128)-2e9) == -2e9 );
+ VERIFY( std::abs((__float128)-3e-4) == 3e-4 );
+}
+
+void
+test_inf()
+{
+ __float128 inf = __builtin_huge_vall();
+ VERIFY( std::abs(inf) == inf );
+ VERIFY( std::abs(-inf) == inf );
+}
+
+#if __cplusplus >= 201103L
+static_assert( std::abs((__float128)-1.0) == (__float128)1.0,
+ "std::abs(__float128) is usable in constant expressions" );
+#endif
+
+int main()
+{
+ test_nan();
+ test_zero();
+}
+#else
+int main() { }
+#endif
reply other threads:[~2024-03-18 14:05 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20240318140521.A97DA385840C@sourceware.org \
--to=redi@gcc.gnu.org \
--cc=gcc-cvs@gcc.gnu.org \
--cc=libstdc++-cvs@gcc.gnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).