From: "H.J. Lu" <hjl.tools@gmail.com>
To: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Cc: libc-alpha@sourceware.org
Subject: Re: [PATCH 06/12] math: Add test to check if round raise inexact floating-point exception
Date: Mon, 1 Apr 2024 07:02:49 -0700 [thread overview]
Message-ID: <CAMe9rOpvWauHtF5-Z9RfickPoOmya_8Q2GY-=6tEo_FxPypZcg@mail.gmail.com> (raw)
In-Reply-To: <20231228172026.2013007-7-adhemerval.zanella@linaro.org>
On Thu, Dec 28, 2023 at 9:22 AM Adhemerval Zanella
<adhemerval.zanella@linaro.org> wrote:
>
> It is not allowed anymore on ISO C23.
> ---
> math/Makefile | 4 ++
> math/test-round-except-2.c | 67 ++++++++++++++++++++++++++++++
> math/test-round-except.c | 85 ++++++++++++++++++++++++++++++++++++++
> 3 files changed, 156 insertions(+)
> create mode 100644 math/test-round-except-2.c
> create mode 100644 math/test-round-except.c
>
> diff --git a/math/Makefile b/math/Makefile
> index 7ae107170c..dfe24a8c03 100644
> --- a/math/Makefile
> +++ b/math/Makefile
> @@ -530,6 +530,8 @@ tests = \
> test-nearbyint-except \
> test-nearbyint-except-2 \
> test-powl \
> + test-round-except \
> + test-round-except-2 \
> test-signgam-uchar \
> test-signgam-uchar-init \
> test-signgam-uint \
> @@ -1001,6 +1003,8 @@ CFLAGS-test-floor-except.c += -fno-builtin
> CFLAGS-test-floor-except-2.c += -fno-builtin
> CFLAGS-test-trunc-except.c += -fno-builtin
> CFLAGS-test-trunc-except-2.c += -fno-builtin
> +CFLAGS-test-round-except.c += -fno-builtin
> +CFLAGS-test-round-except-2.c += -fno-builtin
>
> include ../Rules
>
> diff --git a/math/test-round-except-2.c b/math/test-round-except-2.c
> new file mode 100644
> index 0000000000..dffc02e91f
> --- /dev/null
> +++ b/math/test-round-except-2.c
> @@ -0,0 +1,67 @@
> +/* Test round functions do not disable exception traps.
> + Copyright (C) 2023 Free Software Foundation, Inc.
2024
> + 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/>. */
> +
> +#include <fenv.h>
> +#include <math.h>
> +#include <stdio.h>
> +
> +#ifndef FE_INEXACT
> +# define FE_INEXACT 0
> +#endif
> +
> +#define TEST_FUNC(NAME, FLOAT, SUFFIX) \
> +static int \
> +NAME (void) \
> +{ \
> + int result = 0; \
> + volatile FLOAT a, b __attribute__ ((unused)); \
> + a = 1.5; \
> + /* round must work when traps on "inexact" are enabled. */ \
> + b = round ## SUFFIX (a); \
> + /* And it must have left those traps enabled. */ \
> + if (fegetexcept () == FE_INEXACT) \
> + puts ("PASS: " #FLOAT); \
> + else \
> + { \
> + puts ("FAIL: " #FLOAT); \
> + result = 1; \
> + } \
> + return result; \
> +}
> +
> +TEST_FUNC (float_test, float, f)
> +TEST_FUNC (double_test, double, )
> +TEST_FUNC (ldouble_test, long double, l)
> +
> +static int
> +do_test (void)
> +{
> + if (feenableexcept (FE_INEXACT) == -1)
> + {
> + puts ("enabling FE_INEXACT traps failed, cannot test");
> + return 77;
> + }
> + int result = float_test ();
> + feenableexcept (FE_INEXACT);
> + result |= double_test ();
> + feenableexcept (FE_INEXACT);
> + result |= ldouble_test ();
> + return result;
> +}
> +
> +#include <support/test-driver.c>
> diff --git a/math/test-round-except.c b/math/test-round-except.c
> new file mode 100644
> index 0000000000..cd43c964ab
> --- /dev/null
> +++ b/math/test-round-except.c
> @@ -0,0 +1,85 @@
> +/* Test round functions do not clear exceptions.
> + Copyright (C) 2015-2023 Free Software Foundation, Inc.
Just 2024.
> + 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/>. */
> +
> +#include <fenv.h>
> +#include <math.h>
> +#include <stdbool.h>
> +#include <stdio.h>
> +
> +#include <math-tests.h>
> +
> +#ifndef FE_INVALID
> +# define FE_INVALID 0
> +#endif
> +
> +static bool any_supported = false;
> +
> +#define TEST_FUNC(NAME, FLOAT, SUFFIX) \
> +static int \
> +NAME (void) \
> +{ \
> + int result = 0; \
> + if (!EXCEPTION_TESTS (FLOAT)) \
> + return 0; \
> + any_supported = true; \
> + volatile FLOAT a, b __attribute__ ((unused)); \
> + a = 1.0; \
> + /* round must not clear already-raised exceptions. */ \
> + feraiseexcept (FE_ALL_EXCEPT); \
> + b = round ## SUFFIX (a); \
> + if (fetestexcept (FE_ALL_EXCEPT) == FE_ALL_EXCEPT) \
> + puts ("PASS: " #FLOAT); \
> + else \
> + { \
> + puts ("FAIL: " #FLOAT); \
> + result = 1; \
> + } \
> + /* But it mustn't lose exceptions from sNaN arguments. */ \
> + if (SNAN_TESTS (FLOAT)) \
> + { \
> + static volatile FLOAT snan = __builtin_nans ## SUFFIX (""); \
> + volatile FLOAT c __attribute__ ((unused)); \
> + feclearexcept (FE_ALL_EXCEPT); \
> + c = round ## SUFFIX (snan); \
> + if (fetestexcept (FE_INVALID) == FE_INVALID) \
> + puts ("PASS: " #FLOAT " sNaN"); \
> + else \
> + { \
> + puts ("FAIL: " #FLOAT " sNaN"); \
> + result = 1; \
> + } \
> + } \
> + return result; \
> +}
> +
> +TEST_FUNC (float_test, float, f)
> +TEST_FUNC (double_test, double, )
> +TEST_FUNC (ldouble_test, long double, l)
> +
> +static int
> +do_test (void)
> +{
> + int result = float_test ();
> + result |= double_test ();
> + result |= ldouble_test ();
> + if (!any_supported)
> + return 77;
> + return result;
> +}
> +
> +#include <support/test-driver.c>
> --
> 2.34.1
>
--
H.J.
next prev parent reply other threads:[~2024-04-01 14:04 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-12-28 17:20 [PATCH 00/12] Improve rounding to interger function for C23 Adhemerval Zanella
2023-12-28 17:20 ` [PATCH 01/12] math: Reformat Makefile Adhemerval Zanella
2024-04-01 13:57 ` H.J. Lu
2023-12-28 17:20 ` [PATCH 02/12] powerpc: Add missing arch flags on rounding ifunc variants Adhemerval Zanella
2024-04-01 13:59 ` H.J. Lu
2024-04-01 16:50 ` Peter Bergner
2023-12-28 17:20 ` [PATCH 03/12] math: Add test to check if ceil raise inexact floating-point exception Adhemerval Zanella
2024-04-01 14:01 ` H.J. Lu
2023-12-28 17:20 ` [PATCH 04/12] math: Add test to check if floor " Adhemerval Zanella
2024-04-01 14:02 ` H.J. Lu
2023-12-28 17:20 ` [PATCH 05/12] math: Add test to check if trunc " Adhemerval Zanella
2024-04-01 14:03 ` H.J. Lu
2023-12-28 17:20 ` [PATCH 06/12] math: Add test to check if round " Adhemerval Zanella
2024-04-01 14:02 ` H.J. Lu [this message]
2023-12-28 17:20 ` [PATCH 07/12] x86: Do not raise inexact exception on ceill Adhemerval Zanella
2024-04-01 14:12 ` H.J. Lu
2023-12-28 17:20 ` [PATCH 08/12] x86: Do not raise inexact exception on floorl Adhemerval Zanella
2024-04-01 14:13 ` H.J. Lu
2023-12-28 17:20 ` [PATCH 09/12] x86: Do not raise inexact exception on truncl Adhemerval Zanella
2024-04-01 14:14 ` H.J. Lu
2023-12-28 17:20 ` [PATCH 10/12] x86: Do not raise inexact exception on floor/floorf Adhemerval Zanella
2024-04-01 14:15 ` H.J. Lu
2023-12-28 17:20 ` [PATCH 11/12] i386: Do not raise inexact exception on ceil/ceilf Adhemerval Zanella
2024-04-01 14:16 ` H.J. Lu
2023-12-28 17:20 ` [PATCH 12/12] i386: Do not raise inexact exception on trunc/truncf Adhemerval Zanella
2024-04-01 14:17 ` H.J. Lu
2024-02-16 17:47 ` [PATCH 00/12] Improve rounding to interger function for C23 Adhemerval Zanella Netto
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
List information: https://www.gnu.org/software/libc/involved.html
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CAMe9rOpvWauHtF5-Z9RfickPoOmya_8Q2GY-=6tEo_FxPypZcg@mail.gmail.com' \
--to=hjl.tools@gmail.com \
--cc=adhemerval.zanella@linaro.org \
--cc=libc-alpha@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).