public inbox for newlib-cvs@sourceware.org
help / color / mirror / Atom feed
From: Jeff Johnston <jjohnstn@sourceware.org>
To: newlib-cvs@sourceware.org
Subject: [newlib-cygwin] Fix rounding issues with sqrt/sqrtf
Date: Fri, 4 Jun 2021 18:43:11 +0000 (GMT) [thread overview]
Message-ID: <20210604184311.7059038515F0@sourceware.org> (raw)
https://sourceware.org/git/gitweb.cgi?p=newlib-cygwin.git;h=a9165ea07ce3a99bbfd294ada9c0fd54db738bb9
commit a9165ea07ce3a99bbfd294ada9c0fd54db738bb9
Author: Jeff Johnston <jjohnstn@redhat.com>
Date: Fri Jun 4 14:36:38 2021 -0400
Fix rounding issues with sqrt/sqrtf
- compiler is sometimes optimizing out the rounding check in
e_sqrt.c and ef_sqrt.c which uses two constants to create
an inexact operation
- there is a similar constant operation in s_tanh.c/sf_tanh.c
- make the one and tiny constants volatile to stop this
Diff:
---
newlib/libm/math/e_sqrt.c | 4 ++--
newlib/libm/math/ef_sqrt.c | 4 ++--
newlib/libm/math/s_tanh.c | 4 ++--
newlib/libm/math/sf_tanh.c | 4 ++--
4 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/newlib/libm/math/e_sqrt.c b/newlib/libm/math/e_sqrt.c
index 78fc52417..85dbdd1e7 100644
--- a/newlib/libm/math/e_sqrt.c
+++ b/newlib/libm/math/e_sqrt.c
@@ -86,9 +86,9 @@
#ifndef _DOUBLE_IS_32BITS
#ifdef __STDC__
-static const double one = 1.0, tiny=1.0e-300;
+static const volatile double one = 1.0, tiny=1.0e-300;
#else
-static double one = 1.0, tiny=1.0e-300;
+static double volatile one = 1.0, tiny=1.0e-300;
#endif
#ifdef __STDC__
diff --git a/newlib/libm/math/ef_sqrt.c b/newlib/libm/math/ef_sqrt.c
index 80e7f360e..448e26e57 100644
--- a/newlib/libm/math/ef_sqrt.c
+++ b/newlib/libm/math/ef_sqrt.c
@@ -16,9 +16,9 @@
#include "fdlibm.h"
#ifdef __STDC__
-static const float one = 1.0, tiny=1.0e-30;
+static const volatile float one = 1.0, tiny=1.0e-30;
#else
-static float one = 1.0, tiny=1.0e-30;
+static float volatile one = 1.0, tiny=1.0e-30;
#endif
#ifdef __STDC__
diff --git a/newlib/libm/math/s_tanh.c b/newlib/libm/math/s_tanh.c
index f7864306e..66018d7d0 100644
--- a/newlib/libm/math/s_tanh.c
+++ b/newlib/libm/math/s_tanh.c
@@ -73,9 +73,9 @@ PORTABILITY
#ifndef _DOUBLE_IS_32BITS
#ifdef __STDC__
-static const double one=1.0, two=2.0, tiny = 1.0e-300;
+static const volatile double one=1.0, two=2.0, tiny = 1.0e-300;
#else
-static double one=1.0, two=2.0, tiny = 1.0e-300;
+static double volatile one=1.0, two=2.0, tiny = 1.0e-300;
#endif
#ifdef __STDC__
diff --git a/newlib/libm/math/sf_tanh.c b/newlib/libm/math/sf_tanh.c
index 1eb44a2ff..214bdb495 100644
--- a/newlib/libm/math/sf_tanh.c
+++ b/newlib/libm/math/sf_tanh.c
@@ -16,9 +16,9 @@
#include "fdlibm.h"
#ifdef __STDC__
-static const float one=1.0, two=2.0, tiny = 1.0e-30;
+static const volatile float one=1.0, two=2.0, tiny = 1.0e-30;
#else
-static float one=1.0, two=2.0, tiny = 1.0e-30;
+static volatile float one=1.0, two=2.0, tiny = 1.0e-30;
#endif
#ifdef __STDC__
reply other threads:[~2021-06-04 18:43 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=20210604184311.7059038515F0@sourceware.org \
--to=jjohnstn@sourceware.org \
--cc=newlib-cvs@sourceware.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).