* [PATCH 2/4] Remove finite-math tests
@ 2019-06-26 13:59 Wilco Dijkstra
2019-07-16 18:19 ` Adhemerval Zanella
0 siblings, 1 reply; 3+ messages in thread
From: Wilco Dijkstra @ 2019-06-26 13:59 UTC (permalink / raw)
To: libc-alpha; +Cc: nd
Remove the finite-math tests from the testsuite - these are no longer
useful after removing math-finite.h header.
Passes buildmanyglibc, build&test on x86_64 and AArch64.
2019-06-26 Wilco Dijkstra <wdijkstr@arm.com>
* math/Makefile: Update to remove finite-math tests.
* math/gen-libm-test.py: Remove setting of NON_FINITE.
* math/libm-test-driver.c: Remove TEST_FINITE, flag_test_finite.
* math/libm-test-support.h: Remove flag_test_finite declartion.
* math/libm-test-support.c: Remove NON_FINITE support.
* math/test-finite-macros.c: Remove file.
* math/test-math-finite.h: Likewise.
* math/test-math-inline.h: Remove TEST_FINITE.
* math/test-math-no-finite.h : Remove file.
* math/test-math-vector.h: Remove TEST_FINITE.
* math/test-signgam-finite.c: Remove test.
* math/test-signgam-finite-c11.c: Likewise.
* math/test-signgam-finite-c99.c: Likewise.
--
diff --git a/math/Makefile b/math/Makefile
index 8d754ed02c4e60dc9becc63e2b604e8c0d6557f7..d6eb0c430afd1586a7bcf9ebdb6e3c841dc2839e 100644
--- a/math/Makefile
+++ b/math/Makefile
@@ -228,8 +228,7 @@ tests = test-matherr-3 test-fenv basic-test \
test-tgmath-ret bug-nextafter bug-nexttoward bug-tgmath1 \
test-tgmath-int test-tgmath2 test-powl tst-CMPLX tst-CMPLX2 test-snan \
test-fenv-tls test-fenv-preserve test-fenv-return test-fenvinline \
- test-nearbyint-except test-fenv-clear test-signgam-finite \
- test-signgam-finite-c99 test-signgam-finite-c11 \
+ test-nearbyint-except test-fenv-clear \
test-nearbyint-except-2 test-signgam-uchar test-signgam-uchar-init \
test-signgam-uint test-signgam-uint-init test-signgam-ullong \
test-signgam-ullong-init test-nan-overflow test-nan-payload \
@@ -238,7 +237,7 @@ tests = test-matherr-3 test-fenv basic-test \
test-femode-traps test-iszero-excess-precision \
test-iseqsig-excess-precision test-flt-eval-method \
test-fp-ilogb-constants test-fp-llogb-constants \
- test-fe-snans-always-signal test-finite-macros test-narrow-macros \
+ test-fe-snans-always-signal test-narrow-macros \
test-nan-const $(tests-static)
tests-static = test-fpucw-static test-fpucw-ieee-static \
test-signgam-uchar-static test-signgam-uchar-init-static \
@@ -297,17 +296,13 @@ generated += libm-test-ulps.h $(libm-test-c-auto) $(libm-test-c-noauto) \
$(libm-test-c-narrow)
libm-tests-base-normal = $(foreach t,$(test-types),test-$(t))
-libm-tests-base-finite = $(foreach t,$(test-types),test-$(t)-finite)
libm-tests-base-inline = $(foreach t,$(test-types),test-i$(t))
libm-tests-base-narrow = $(foreach t,$(test-type-pairs),test-$(t))
-libm-tests-base = $(libm-tests-base-normal) $(libm-tests-base-finite) \
- $(libm-tests-base-inline) $(libm-vec-tests)
+libm-tests-base = $(libm-tests-base-normal) $(libm-tests-base-inline) \
+ $(libm-vec-tests)
libm-tests-normal = $(foreach t,$(libm-tests-base-normal),\
$(foreach f,$(libm-test-funcs-all),\
$(t)-$(f)))
-libm-tests-finite = $(foreach t,$(libm-tests-base-finite),\
- $(foreach f,$(libm-test-funcs-all),\
- $(t)-$(f)))
libm-tests-inline = $(foreach t,$(libm-tests-base-inline),\
$(foreach f,$(libm-test-funcs-all),\
$(t)-$(f)))
@@ -316,11 +311,10 @@ libm-tests-narrow = $(foreach t,$(libm-tests-base-narrow),\
$(t)-$(f)))
libm-tests-vector = $(foreach t,$(libmvec-tests),\
$(foreach f,$($(t)-funcs),test-$(t)-$(f)))
-libm-tests = $(libm-tests-normal) $(libm-tests-finite) $(libm-tests-inline) \
+libm-tests = $(libm-tests-normal) $(libm-tests-inline) \
$(libm-tests-narrow) $(libm-tests-vector)
libm-tests-for-type = $(foreach f,$(libm-test-funcs-all),\
- test-$(1)-$(f) test-$(1)-finite-$(f) \
- test-i$(1)-$(f)) \
+ test-$(1)-$(f) test-i$(1)-$(f)) \
$(filter test-$(1)-%,$(libm-tests-vector) \
$(libm-tests-narrow))
@@ -398,8 +392,6 @@ CFLAGS-test-float-vlen16-wrappers.c += $(float-vlen16-arch-ext-cflags)
# validated.
libm-test-no-inline-cflags = -fno-inline -ffloat-store -fno-builtin \
-fsignaling-nans
-libm-test-finite-cflags = $(libm-test-no-inline-cflags) \
- -U__FINITE_MATH_ONLY__ -D__FINITE_MATH_ONLY__=1
libm-test-inline-cflags = $(libm-test-fast-math-cflags)
CFLAGS-test-tgmath.c += -fno-builtin
# The following testcase uses very long lines (>3 million), so it may take a
@@ -411,9 +403,6 @@ CFLAGS-test-powl.c += -fno-builtin
CFLAGS-test-snan.c += -fsignaling-nans
-CFLAGS-test-signgam-finite.c += -ffinite-math-only
-CFLAGS-test-signgam-finite-c99.c += -ffinite-math-only -std=c99
-CFLAGS-test-signgam-finite-c11.c += -ffinite-math-only -std=c11
CFLAGS-test-signgam-uchar.c += -std=c99
CFLAGS-test-signgam-uchar-init.c += -std=c99
CFLAGS-test-signgam-uchar-static.c += -std=c99
@@ -440,8 +429,6 @@ CFLAGS-test-flt-eval-method.c += -fexcess-precision=standard
CFLAGS-test-fe-snans-always-signal.c += -fsignaling-nans
-CFLAGS-test-finite-macros.c += -ffinite-math-only
-
CFLAGS-test-nan-const.c += -fno-builtin
include ../Rules
@@ -480,25 +467,13 @@ $(foreach t,$(libm-tests-normal),$(objpfx)$(t).c): $(objpfx)test-%.c:
func=$${type_func#*-}; \
( \
echo "#include <test-$$type.h>"; \
- echo "#include <test-math-no-finite.h>"; \
+ echo "#define TEST_EXCEPTIONS 1"; \
echo "#include <test-math-no-inline.h>"; \
echo "#include <test-math-errno.h>"; \
echo "#include <test-math-scalar.h>"; \
echo "#include <libm-test-$$func.c>"; \
) > $@
-$(foreach t,$(libm-tests-finite),$(objpfx)$(t).c): $(objpfx)test-%.c:
- type_func=$*; \
- type=$${type_func%%-*}; \
- func=$${type_func#*-finite-}; \
- ( \
- echo "#include <test-$$type.h>"; \
- echo "#include <test-math-finite.h>"; \
- echo "#include <test-math-no-inline.h>"; \
- echo "#include <test-math-scalar.h>"; \
- echo "#include <libm-test-$$func.c>"; \
- ) > $@
-
$(foreach t,$(libm-tests-inline),$(objpfx)$(t).c): $(objpfx)test-i%.c:
type_func=$*; \
type=$${type_func%%-*}; \
@@ -519,7 +494,7 @@ $(foreach t,$(libm-tests-narrow),$(objpfx)$(t).c): $(objpfx)test-%.c:
( \
echo "#include <test-$$ret_type.h>"; \
echo "#include <test-arg-$$arg_type.h>"; \
- echo "#include <test-math-no-finite.h>"; \
+ echo "#define TEST_EXCEPTIONS 1"; \
echo "#include <test-math-no-inline.h>"; \
echo "#include <test-math-errno.h>"; \
echo "#include <test-math-narrow.h>"; \
@@ -571,13 +546,6 @@ include $(o-iterator)
define o-iterator-doit
$(foreach f,$(libm-test-funcs-all),\
- $(objpfx)$(o)-$(f).o): CFLAGS += $(libm-test-finite-cflags)
-endef
-object-suffixes-left := $(libm-tests-base-finite)
-include $(o-iterator)
-
-define o-iterator-doit
-$(foreach f,$(libm-test-funcs-all),\
$(objpfx)$(o)-$(f).o): CFLAGS += $(libm-test-inline-cflags)
endef
object-suffixes-left := $(libm-tests-base-inline)
diff --git a/math/gen-libm-test.py b/math/gen-libm-test.py
index 4574215fec7b385b0513f85622f3cdd69c8171ab..b98a69622f5c6988dfa5ccd60e5bf9c2d271849f 100755
--- a/math/gen-libm-test.py
+++ b/math/gen-libm-test.py
@@ -341,9 +341,6 @@ def apply_lit(arg, macro):
def gen_test_args_res(descr_args, descr_res, args, res_rm):
"""Generate a test given the arguments and per-rounding-mode results."""
- # Determine whether any arguments or results, for any rounding
- # mode, are non-finite.
- non_finite = False
test_snan = False
all_args_res = list(args)
for r in res_rm:
@@ -351,10 +348,6 @@ def gen_test_args_res(descr_args, descr_res, args, res_rm):
for a in all_args_res:
if 'snan_value' in a:
test_snan = True
- non_finite = True
- elif ('qnan_value' in a or 'plus_infty' in a or 'minus_infty' in a
- or 'plus_oflow' in a or 'minus_oflow' in a):
- non_finite = True
# Process the arguments.
args_disp = []
args_c = []
@@ -422,8 +415,6 @@ def gen_test_args_res(descr_args, descr_res, args, res_rm):
flags = []
if ignore_result_any:
flags.append('IGNORE_RESULT')
- if non_finite:
- flags.append('NON_FINITE')
if test_snan:
flags.append('TEST_SNAN')
flags.append(res[res_pos])
diff --git a/math/libm-test-driver.c b/math/libm-test-driver.c
index 4640f30d64ae800687c6fe737b43c0d305847eb2..7623df549f86775b85194bbf60ec685cd4a2745f 100644
--- a/math/libm-test-driver.c
+++ b/math/libm-test-driver.c
@@ -23,7 +23,6 @@
/* Flags set by the including file. */
const int flag_test_errno = TEST_ERRNO;
const int flag_test_exceptions = TEST_EXCEPTIONS;
-const int flag_test_finite = TEST_FINITE;
const int flag_test_inline = TEST_INLINE;
const int flag_test_mathvec = TEST_MATHVEC;
@@ -44,8 +43,6 @@ const int snan_tests_arg = SNAN_TESTS (FLOAT);
# define TEST_MSG "testing " STR_FLOAT " (vector length " STR_VEC_LEN ")\n"
#elif TEST_INLINE
# define TEST_MSG "testing " STR_FLOAT " (inline functions)\n"
-#elif TEST_FINITE
-# define TEST_MSG "testing " STR_FLOAT " (finite-math-only)\n"
#elif TEST_NARROW
# define TEST_MSG "testing " STR_FLOAT " (argument " STR_ARG_FLOAT ")\n"
#else
diff --git a/math/libm-test-support.h b/math/libm-test-support.h
index 2b8d5a3d889a76cc1e678d5db190944d84058cc4..028cf5647027ca5c9c8c7e70f33bd2ea51871afd 100644
--- a/math/libm-test-support.h
+++ b/math/libm-test-support.h
@@ -34,7 +34,6 @@
extern const int flag_test_errno;
extern const int flag_test_exceptions;
-extern const int flag_test_finite;
extern const int flag_test_inline;
extern const int flag_test_mathvec;
extern const int snan_tests_arg;
@@ -69,7 +68,6 @@ extern const char doc[];
#define ERRNO_ERANGE 0x20000
/* Flags generated by gen-libm-test.py, not entered here manually. */
#define IGNORE_RESULT 0x40000
-#define NON_FINITE 0x80000
#define TEST_SNAN 0x100000
#define NO_TEST_MATHVEC 0x200000
diff --git a/math/libm-test-support.c b/math/libm-test-support.c
index 5f986da75229746e62368583d4d6d8602b86d25b..ca70722720f0b6a8fb7a2e4194e7d419e5ac047c 100644
--- a/math/libm-test-support.c
+++ b/math/libm-test-support.c
@@ -983,8 +983,6 @@ enable_test (int exceptions)
return 0;
if (flag_test_inline && (exceptions & NO_TEST_INLINE))
return 0;
- if (flag_test_finite && (exceptions & NON_FINITE) != 0)
- return 0;
if ((!SNAN_TESTS (FLOAT) || !snan_tests_arg)
&& (exceptions & TEST_SNAN) != 0)
return 0;
diff --git a/math/test-finite-macros.c b/math/test-finite-macros.c
deleted file mode 100644
index 14524a357b837d020f48aec10d05ed9ca1855c4f..0000000000000000000000000000000000000000
--- a/math/test-finite-macros.c
+++ /dev/null
@@ -1,35 +0,0 @@
-/* Test finite-math-only code does not conflict with user macros (bug 22028).
- Copyright (C) 2017-2019 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/>. */
-
-/* The main test is that the inclusion of <math.h> compiles. */
-#define f first test macro
-#define l second test macro
-#define f128 third test macro
-
-#include <math.h>
-
-volatile float a, b;
-
-static int
-do_test (void)
-{
- b = acosf (a);
- return 0;
-}
-
-#include <support/test-driver.c>
diff --git a/math/test-math-finite.h b/math/test-math-finite.h
deleted file mode 100644
index e3c8e0a2b9f9aa547c8af0d8a38b26c6830ab6ca..0000000000000000000000000000000000000000
--- a/math/test-math-finite.h
+++ /dev/null
@@ -1,21 +0,0 @@
-/* Common definitions for libm tests for finite-math-only.
- Copyright (C) 2015-2019 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/>. */
-
-#define TEST_FINITE 1
-#define TEST_ERRNO 0
-#define TEST_EXCEPTIONS 1
diff --git a/math/test-math-inline.h b/math/test-math-inline.h
index b51cf52901b693aba3088da3774964c3b852e912..14553dda6ca6e82bdcf42b6122d8a65a9dd06dba 100644
--- a/math/test-math-inline.h
+++ b/math/test-math-inline.h
@@ -17,7 +17,6 @@
<http://www.gnu.org/licenses/>. */
#define TEST_INLINE 1
-#define TEST_FINITE 0
#define TEST_ERRNO 0
#define TEST_EXCEPTIONS 0
diff --git a/math/test-math-no-finite.h b/math/test-math-no-finite.h
deleted file mode 100644
index e3456aae756375f41ef7d4cf47bb68a24dfdddfc..0000000000000000000000000000000000000000
--- a/math/test-math-no-finite.h
+++ /dev/null
@@ -1,20 +0,0 @@
-/* Common definitions for libm tests for no-finite-math-only.
- Copyright (C) 2015-2019 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/>. */
-
-#define TEST_FINITE 0
-#define TEST_EXCEPTIONS 1
diff --git a/math/test-math-vector.h b/math/test-math-vector.h
index b211ddebec08daafb76a8673dbfc7bf11659c0ac..775198867e097f0f8b1e3266aba2eba5a75e94a7 100644
--- a/math/test-math-vector.h
+++ b/math/test-math-vector.h
@@ -18,7 +18,6 @@
#define TEST_MATHVEC 1
#define TEST_NARROW 0
-#define TEST_FINITE 0
#define TEST_ERRNO 0
#define TEST_EXCEPTIONS 0
diff --git a/math/test-signgam-finite-c11.c b/math/test-signgam-finite-c11.c
deleted file mode 100644
index dff132df934fdb2ceb593ed9326e821b4349bd8f..0000000000000000000000000000000000000000
--- a/math/test-signgam-finite-c11.c
+++ /dev/null
@@ -1 +0,0 @@
-#include "test-signgam-finite-c99.c"
diff --git a/math/test-signgam-finite-c99.c b/math/test-signgam-finite-c99.c
deleted file mode 100644
index 9be03a9edc1222953b3f5c331fb729f3dfab3335..0000000000000000000000000000000000000000
--- a/math/test-signgam-finite-c99.c
+++ /dev/null
@@ -1,63 +0,0 @@
-/* Test lgamma functions do not set signgam for -ffinite-math-only for ISO C.
- Copyright (C) 2015-2019 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/>. */
-
-#undef _LIBC
-#undef __LIBC_INTERNAL_MATH_INLINES
-#undef _GNU_SOURCE
-
-#include <math.h>
-#include <stdio.h>
-
-int signgam;
-
-#define RUN_TESTS(FUNC, TYPE) \
- do \
- { \
- volatile TYPE a, b, c __attribute__ ((unused)); \
- a = 0.5; \
- b = -0.5; \
- signgam = 123; \
- c = FUNC (a); \
- if (signgam == 123) \
- puts ("PASS: " #FUNC " (0.5) setting signgam"); \
- else \
- { \
- puts ("FAIL: " #FUNC " (0.5) setting signgam"); \
- result = 1; \
- } \
- signgam = 123; \
- c = FUNC (b); \
- if (signgam == 123) \
- puts ("PASS: " #FUNC " (-0.5) setting signgam"); \
- else \
- { \
- puts ("FAIL: " #FUNC " (-0.5) setting signgam"); \
- result = 1; \
- } \
- } \
- while (0)
-
-int
-main (void)
-{
- int result = 0;
- RUN_TESTS (lgammaf, float);
- RUN_TESTS (lgamma, double);
- RUN_TESTS (lgammal, long double);
- return result;
-}
diff --git a/math/test-signgam-finite.c b/math/test-signgam-finite.c
deleted file mode 100644
index 5ec697c7738d81b87099d9585ac14b58b3d091c2..0000000000000000000000000000000000000000
--- a/math/test-signgam-finite.c
+++ /dev/null
@@ -1,63 +0,0 @@
-/* Test lgamma functions set signgam for -ffinite-math-only (bug 19211).
- Copyright (C) 2015-2019 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 <math.h>
-#include <stdio.h>
-
-#define RUN_TESTS(FUNC, TYPE) \
- do \
- { \
- volatile TYPE a, b, c __attribute__ ((unused)); \
- a = 0.5; \
- b = -0.5; \
- signgam = 123; \
- c = FUNC (a); \
- if (signgam == 1) \
- puts ("PASS: " #FUNC " (0.5) setting signgam"); \
- else \
- { \
- puts ("FAIL: " #FUNC " (0.5) setting signgam"); \
- result = 1; \
- } \
- signgam = 123; \
- c = FUNC (b); \
- if (signgam == -1) \
- puts ("PASS: " #FUNC " (-0.5) setting signgam"); \
- else \
- { \
- puts ("FAIL: " #FUNC " (-0.5) setting signgam"); \
- result = 1; \
- } \
- } \
- while (0)
-
-static int
-do_test (void)
-{
- int result = 0;
- RUN_TESTS (lgammaf, float);
- RUN_TESTS (gammaf, float);
- RUN_TESTS (lgamma, double);
- RUN_TESTS (gamma, double);
- RUN_TESTS (lgammal, long double);
- RUN_TESTS (gammal, long double);
- return result;
-}
-
-#define TEST_FUNCTION do_test ()
-#include "../test-skeleton.c"
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH 2/4] Remove finite-math tests
2019-06-26 13:59 [PATCH 2/4] Remove finite-math tests Wilco Dijkstra
@ 2019-07-16 18:19 ` Adhemerval Zanella
2019-10-18 18:14 ` Adhemerval Zanella
0 siblings, 1 reply; 3+ messages in thread
From: Adhemerval Zanella @ 2019-07-16 18:19 UTC (permalink / raw)
To: Wilco Dijkstra, libc-alpha; +Cc: nd
On 26/06/2019 10:59, Wilco Dijkstra wrote:
> Remove the finite-math tests from the testsuite - these are no longer
> useful after removing math-finite.h header.
>
> Passes buildmanyglibc, build&test on x86_64 and AArch64.
>
> 2019-06-26 Wilco Dijkstra <wdijkstr@arm.com>
>
> * math/Makefile: Update to remove finite-math tests.
> * math/gen-libm-test.py: Remove setting of NON_FINITE.
> * math/libm-test-driver.c: Remove TEST_FINITE, flag_test_finite.
> * math/libm-test-support.h: Remove flag_test_finite declartion.
s/declartion/declaration
> * math/libm-test-support.c: Remove NON_FINITE support.
> * math/test-finite-macros.c: Remove file.
> * math/test-math-finite.h: Likewise.
> * math/test-math-inline.h: Remove TEST_FINITE.
> * math/test-math-no-finite.h : Remove file.
> * math/test-math-vector.h: Remove TEST_FINITE.
> * math/test-signgam-finite.c: Remove test.
> * math/test-signgam-finite-c11.c: Likewise.
> * math/test-signgam-finite-c99.c: Likewise.
LGTM with just one modification regarding TEST_EXCEPTIONS below.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
>
> --
> diff --git a/math/Makefile b/math/Makefile
> index 8d754ed02c4e60dc9becc63e2b604e8c0d6557f7..d6eb0c430afd1586a7bcf9ebdb6e3c841dc2839e 100644
> --- a/math/Makefile
> +++ b/math/Makefile
> @@ -228,8 +228,7 @@ tests = test-matherr-3 test-fenv basic-test \
> test-tgmath-ret bug-nextafter bug-nexttoward bug-tgmath1 \
> test-tgmath-int test-tgmath2 test-powl tst-CMPLX tst-CMPLX2 test-snan \
> test-fenv-tls test-fenv-preserve test-fenv-return test-fenvinline \
> - test-nearbyint-except test-fenv-clear test-signgam-finite \
> - test-signgam-finite-c99 test-signgam-finite-c11 \
> + test-nearbyint-except test-fenv-clear \
> test-nearbyint-except-2 test-signgam-uchar test-signgam-uchar-init \
> test-signgam-uint test-signgam-uint-init test-signgam-ullong \
> test-signgam-ullong-init test-nan-overflow test-nan-payload \
> @@ -238,7 +237,7 @@ tests = test-matherr-3 test-fenv basic-test \
> test-femode-traps test-iszero-excess-precision \
> test-iseqsig-excess-precision test-flt-eval-method \
> test-fp-ilogb-constants test-fp-llogb-constants \
> - test-fe-snans-always-signal test-finite-macros test-narrow-macros \
> + test-fe-snans-always-signal test-narrow-macros \
> test-nan-const $(tests-static)
> tests-static = test-fpucw-static test-fpucw-ieee-static \
> test-signgam-uchar-static test-signgam-uchar-init-static \
Ok.
> @@ -297,17 +296,13 @@ generated += libm-test-ulps.h $(libm-test-c-auto) $(libm-test-c-noauto) \
> $(libm-test-c-narrow)
>
> libm-tests-base-normal = $(foreach t,$(test-types),test-$(t))
> -libm-tests-base-finite = $(foreach t,$(test-types),test-$(t)-finite)
> libm-tests-base-inline = $(foreach t,$(test-types),test-i$(t))
> libm-tests-base-narrow = $(foreach t,$(test-type-pairs),test-$(t))
> -libm-tests-base = $(libm-tests-base-normal) $(libm-tests-base-finite) \
> - $(libm-tests-base-inline) $(libm-vec-tests)
> +libm-tests-base = $(libm-tests-base-normal) $(libm-tests-base-inline) \
> + $(libm-vec-tests)
> libm-tests-normal = $(foreach t,$(libm-tests-base-normal),\
> $(foreach f,$(libm-test-funcs-all),\
> $(t)-$(f)))
> -libm-tests-finite = $(foreach t,$(libm-tests-base-finite),\
> - $(foreach f,$(libm-test-funcs-all),\
> - $(t)-$(f)))
> libm-tests-inline = $(foreach t,$(libm-tests-base-inline),\
> $(foreach f,$(libm-test-funcs-all),\
> $(t)-$(f)))
Ok.
> @@ -316,11 +311,10 @@ libm-tests-narrow = $(foreach t,$(libm-tests-base-narrow),\
> $(t)-$(f)))
> libm-tests-vector = $(foreach t,$(libmvec-tests),\
> $(foreach f,$($(t)-funcs),test-$(t)-$(f)))
> -libm-tests = $(libm-tests-normal) $(libm-tests-finite) $(libm-tests-inline) \
> +libm-tests = $(libm-tests-normal) $(libm-tests-inline) \
> $(libm-tests-narrow) $(libm-tests-vector)
> libm-tests-for-type = $(foreach f,$(libm-test-funcs-all),\
> - test-$(1)-$(f) test-$(1)-finite-$(f) \
> - test-i$(1)-$(f)) \
> + test-$(1)-$(f) test-i$(1)-$(f)) \
> $(filter test-$(1)-%,$(libm-tests-vector) \
> $(libm-tests-narrow))
>
Ok.
> @@ -398,8 +392,6 @@ CFLAGS-test-float-vlen16-wrappers.c += $(float-vlen16-arch-ext-cflags)
> # validated.
> libm-test-no-inline-cflags = -fno-inline -ffloat-store -fno-builtin \
> -fsignaling-nans
> -libm-test-finite-cflags = $(libm-test-no-inline-cflags) \
> - -U__FINITE_MATH_ONLY__ -D__FINITE_MATH_ONLY__=1
> libm-test-inline-cflags = $(libm-test-fast-math-cflags)
> CFLAGS-test-tgmath.c += -fno-builtin
> # The following testcase uses very long lines (>3 million), so it may take a
Ok.
> @@ -411,9 +403,6 @@ CFLAGS-test-powl.c += -fno-builtin
>
> CFLAGS-test-snan.c += -fsignaling-nans
>
> -CFLAGS-test-signgam-finite.c += -ffinite-math-only
> -CFLAGS-test-signgam-finite-c99.c += -ffinite-math-only -std=c99
> -CFLAGS-test-signgam-finite-c11.c += -ffinite-math-only -std=c11
> CFLAGS-test-signgam-uchar.c += -std=c99
> CFLAGS-test-signgam-uchar-init.c += -std=c99
> CFLAGS-test-signgam-uchar-static.c += -std=c99
Ok.
> @@ -440,8 +429,6 @@ CFLAGS-test-flt-eval-method.c += -fexcess-precision=standard
>
> CFLAGS-test-fe-snans-always-signal.c += -fsignaling-nans
>
> -CFLAGS-test-finite-macros.c += -ffinite-math-only
> -
> CFLAGS-test-nan-const.c += -fno-builtin
>
> include ../Rules
Ok.
> @@ -480,25 +467,13 @@ $(foreach t,$(libm-tests-normal),$(objpfx)$(t).c): $(objpfx)test-%.c:
> func=$${type_func#*-}; \
> ( \
> echo "#include <test-$$type.h>"; \
> - echo "#include <test-math-no-finite.h>"; \
> + echo "#define TEST_EXCEPTIONS 1"; \
> echo "#include <test-math-no-inline.h>"; \
> echo "#include <test-math-errno.h>"; \
> echo "#include <test-math-scalar.h>"; \
> echo "#include <libm-test-$$func.c>"; \
> ) > $@
>
I think we should follow the current organization and create a test-math-exceptions.h
instead of defining on the makefile rule.
> -$(foreach t,$(libm-tests-finite),$(objpfx)$(t).c): $(objpfx)test-%.c:
> - type_func=$*; \
> - type=$${type_func%%-*}; \
> - func=$${type_func#*-finite-}; \
> - ( \
> - echo "#include <test-$$type.h>"; \
> - echo "#include <test-math-finite.h>"; \
> - echo "#include <test-math-no-inline.h>"; \
> - echo "#include <test-math-scalar.h>"; \
> - echo "#include <libm-test-$$func.c>"; \
> - ) > $@
> -
> $(foreach t,$(libm-tests-inline),$(objpfx)$(t).c): $(objpfx)test-i%.c:
> type_func=$*; \
> type=$${type_func%%-*}; \
Ok.
> @@ -519,7 +494,7 @@ $(foreach t,$(libm-tests-narrow),$(objpfx)$(t).c): $(objpfx)test-%.c:
> ( \
> echo "#include <test-$$ret_type.h>"; \
> echo "#include <test-arg-$$arg_type.h>"; \
> - echo "#include <test-math-no-finite.h>"; \
> + echo "#define TEST_EXCEPTIONS 1"; \
> echo "#include <test-math-no-inline.h>"; \
> echo "#include <test-math-errno.h>"; \
> echo "#include <test-math-narrow.h>"; \
> @@ -571,13 +546,6 @@ include $(o-iterator)
>
> define o-iterator-doit
> $(foreach f,$(libm-test-funcs-all),\
> - $(objpfx)$(o)-$(f).o): CFLAGS += $(libm-test-finite-cflags)
> -endef
> -object-suffixes-left := $(libm-tests-base-finite)
> -include $(o-iterator)
> -
> -define o-iterator-doit
> -$(foreach f,$(libm-test-funcs-all),\
> $(objpfx)$(o)-$(f).o): CFLAGS += $(libm-test-inline-cflags)
> endef
> object-suffixes-left := $(libm-tests-base-inline)
Ok.
> diff --git a/math/gen-libm-test.py b/math/gen-libm-test.py
> index 4574215fec7b385b0513f85622f3cdd69c8171ab..b98a69622f5c6988dfa5ccd60e5bf9c2d271849f 100755
> --- a/math/gen-libm-test.py
> +++ b/math/gen-libm-test.py
> @@ -341,9 +341,6 @@ def apply_lit(arg, macro):
>
> def gen_test_args_res(descr_args, descr_res, args, res_rm):
> """Generate a test given the arguments and per-rounding-mode results."""
> - # Determine whether any arguments or results, for any rounding
> - # mode, are non-finite.
> - non_finite = False
> test_snan = False
> all_args_res = list(args)
> for r in res_rm:
> @@ -351,10 +348,6 @@ def gen_test_args_res(descr_args, descr_res, args, res_rm):
> for a in all_args_res:
> if 'snan_value' in a:
> test_snan = True
> - non_finite = True
> - elif ('qnan_value' in a or 'plus_infty' in a or 'minus_infty' in a
> - or 'plus_oflow' in a or 'minus_oflow' in a):
> - non_finite = True
> # Process the arguments.
> args_disp = []
> args_c = []
> @@ -422,8 +415,6 @@ def gen_test_args_res(descr_args, descr_res, args, res_rm):
> flags = []
> if ignore_result_any:
> flags.append('IGNORE_RESULT')
> - if non_finite:
> - flags.append('NON_FINITE')
> if test_snan:
> flags.append('TEST_SNAN')
> flags.append(res[res_pos])
Ok.
> diff --git a/math/libm-test-driver.c b/math/libm-test-driver.c
> index 4640f30d64ae800687c6fe737b43c0d305847eb2..7623df549f86775b85194bbf60ec685cd4a2745f 100644
> --- a/math/libm-test-driver.c
> +++ b/math/libm-test-driver.c
> @@ -23,7 +23,6 @@
> /* Flags set by the including file. */
> const int flag_test_errno = TEST_ERRNO;
> const int flag_test_exceptions = TEST_EXCEPTIONS;
> -const int flag_test_finite = TEST_FINITE;
> const int flag_test_inline = TEST_INLINE;
> const int flag_test_mathvec = TEST_MATHVEC;
>
> @@ -44,8 +43,6 @@ const int snan_tests_arg = SNAN_TESTS (FLOAT);
> # define TEST_MSG "testing " STR_FLOAT " (vector length " STR_VEC_LEN ")\n"
> #elif TEST_INLINE
> # define TEST_MSG "testing " STR_FLOAT " (inline functions)\n"
> -#elif TEST_FINITE
> -# define TEST_MSG "testing " STR_FLOAT " (finite-math-only)\n"
> #elif TEST_NARROW
> # define TEST_MSG "testing " STR_FLOAT " (argument " STR_ARG_FLOAT ")\n"
> #else
Ok.
> diff --git a/math/libm-test-support.h b/math/libm-test-support.h
> index 2b8d5a3d889a76cc1e678d5db190944d84058cc4..028cf5647027ca5c9c8c7e70f33bd2ea51871afd 100644
> --- a/math/libm-test-support.h
> +++ b/math/libm-test-support.h
> @@ -34,7 +34,6 @@
>
> extern const int flag_test_errno;
> extern const int flag_test_exceptions;
> -extern const int flag_test_finite;
> extern const int flag_test_inline;
> extern const int flag_test_mathvec;
> extern const int snan_tests_arg;
> @@ -69,7 +68,6 @@ extern const char doc[];
> #define ERRNO_ERANGE 0x20000
> /* Flags generated by gen-libm-test.py, not entered here manually. */
> #define IGNORE_RESULT 0x40000
> -#define NON_FINITE 0x80000
> #define TEST_SNAN 0x100000
> #define NO_TEST_MATHVEC 0x200000
>
Ok.
> diff --git a/math/libm-test-support.c b/math/libm-test-support.c
> index 5f986da75229746e62368583d4d6d8602b86d25b..ca70722720f0b6a8fb7a2e4194e7d419e5ac047c 100644
> --- a/math/libm-test-support.c
> +++ b/math/libm-test-support.c
> @@ -983,8 +983,6 @@ enable_test (int exceptions)
> return 0;
> if (flag_test_inline && (exceptions & NO_TEST_INLINE))
> return 0;
> - if (flag_test_finite && (exceptions & NON_FINITE) != 0)
> - return 0;
> if ((!SNAN_TESTS (FLOAT) || !snan_tests_arg)
> && (exceptions & TEST_SNAN) != 0)
> return 0;
Ok.
> diff --git a/math/test-finite-macros.c b/math/test-finite-macros.c
> deleted file mode 100644
> index 14524a357b837d020f48aec10d05ed9ca1855c4f..0000000000000000000000000000000000000000
> --- a/math/test-finite-macros.c
> +++ /dev/null
> @@ -1,35 +0,0 @@
> -/* Test finite-math-only code does not conflict with user macros (bug 22028).
> - Copyright (C) 2017-2019 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/>. */
> -
> -/* The main test is that the inclusion of <math.h> compiles. */
> -#define f first test macro
> -#define l second test macro
> -#define f128 third test macro
> -
> -#include <math.h>
> -
> -volatile float a, b;
> -
> -static int
> -do_test (void)
> -{
> - b = acosf (a);
> - return 0;
> -}
> -
> -#include <support/test-driver.c>
Ok.
> diff --git a/math/test-math-finite.h b/math/test-math-finite.h
> deleted file mode 100644
> index e3c8e0a2b9f9aa547c8af0d8a38b26c6830ab6ca..0000000000000000000000000000000000000000
> --- a/math/test-math-finite.h
> +++ /dev/null
> @@ -1,21 +0,0 @@
> -/* Common definitions for libm tests for finite-math-only.
> - Copyright (C) 2015-2019 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/>. */
> -
> -#define TEST_FINITE 1
> -#define TEST_ERRNO 0
> -#define TEST_EXCEPTIONS 1
Ok.
> diff --git a/math/test-math-inline.h b/math/test-math-inline.h
> index b51cf52901b693aba3088da3774964c3b852e912..14553dda6ca6e82bdcf42b6122d8a65a9dd06dba 100644
> --- a/math/test-math-inline.h
> +++ b/math/test-math-inline.h
> @@ -17,7 +17,6 @@
> <http://www.gnu.org/licenses/>. */
>
> #define TEST_INLINE 1
> -#define TEST_FINITE 0
> #define TEST_ERRNO 0
> #define TEST_EXCEPTIONS 0
>
Ok.
> diff --git a/math/test-math-no-finite.h b/math/test-math-no-finite.h
> deleted file mode 100644
> index e3456aae756375f41ef7d4cf47bb68a24dfdddfc..0000000000000000000000000000000000000000
> --- a/math/test-math-no-finite.h
> +++ /dev/null
> @@ -1,20 +0,0 @@
> -/* Common definitions for libm tests for no-finite-math-only.
> - Copyright (C) 2015-2019 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/>. */
> -
> -#define TEST_FINITE 0
> -#define TEST_EXCEPTIONS 1
Ok.
> diff --git a/math/test-math-vector.h b/math/test-math-vector.h
> index b211ddebec08daafb76a8673dbfc7bf11659c0ac..775198867e097f0f8b1e3266aba2eba5a75e94a7 100644
> --- a/math/test-math-vector.h
> +++ b/math/test-math-vector.h
> @@ -18,7 +18,6 @@
>
> #define TEST_MATHVEC 1
> #define TEST_NARROW 0
> -#define TEST_FINITE 0
> #define TEST_ERRNO 0
> #define TEST_EXCEPTIONS 0
>
Ok.
> diff --git a/math/test-signgam-finite-c11.c b/math/test-signgam-finite-c11.c
> deleted file mode 100644
> index dff132df934fdb2ceb593ed9326e821b4349bd8f..0000000000000000000000000000000000000000
> --- a/math/test-signgam-finite-c11.c
> +++ /dev/null
> @@ -1 +0,0 @@
> -#include "test-signgam-finite-c99.c"
Ok.
> diff --git a/math/test-signgam-finite-c99.c b/math/test-signgam-finite-c99.c
> deleted file mode 100644
> index 9be03a9edc1222953b3f5c331fb729f3dfab3335..0000000000000000000000000000000000000000
> --- a/math/test-signgam-finite-c99.c
> +++ /dev/null
> @@ -1,63 +0,0 @@
> -/* Test lgamma functions do not set signgam for -ffinite-math-only for ISO C.
> - Copyright (C) 2015-2019 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/>. */
> -
> -#undef _LIBC
> -#undef __LIBC_INTERNAL_MATH_INLINES
> -#undef _GNU_SOURCE
> -
> -#include <math.h>
> -#include <stdio.h>
> -
> -int signgam;
> -
> -#define RUN_TESTS(FUNC, TYPE) \
> - do \
> - { \
> - volatile TYPE a, b, c __attribute__ ((unused)); \
> - a = 0.5; \
> - b = -0.5; \
> - signgam = 123; \
> - c = FUNC (a); \
> - if (signgam == 123) \
> - puts ("PASS: " #FUNC " (0.5) setting signgam"); \
> - else \
> - { \
> - puts ("FAIL: " #FUNC " (0.5) setting signgam"); \
> - result = 1; \
> - } \
> - signgam = 123; \
> - c = FUNC (b); \
> - if (signgam == 123) \
> - puts ("PASS: " #FUNC " (-0.5) setting signgam"); \
> - else \
> - { \
> - puts ("FAIL: " #FUNC " (-0.5) setting signgam"); \
> - result = 1; \
> - } \
> - } \
> - while (0)
> -
> -int
> -main (void)
> -{
> - int result = 0;
> - RUN_TESTS (lgammaf, float);
> - RUN_TESTS (lgamma, double);
> - RUN_TESTS (lgammal, long double);
> - return result;
> -}
Ok.
> diff --git a/math/test-signgam-finite.c b/math/test-signgam-finite.c
> deleted file mode 100644
> index 5ec697c7738d81b87099d9585ac14b58b3d091c2..0000000000000000000000000000000000000000
> --- a/math/test-signgam-finite.c
> +++ /dev/null
> @@ -1,63 +0,0 @@
> -/* Test lgamma functions set signgam for -ffinite-math-only (bug 19211).
> - Copyright (C) 2015-2019 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 <math.h>
> -#include <stdio.h>
> -
> -#define RUN_TESTS(FUNC, TYPE) \
> - do \
> - { \
> - volatile TYPE a, b, c __attribute__ ((unused)); \
> - a = 0.5; \
> - b = -0.5; \
> - signgam = 123; \
> - c = FUNC (a); \
> - if (signgam == 1) \
> - puts ("PASS: " #FUNC " (0.5) setting signgam"); \
> - else \
> - { \
> - puts ("FAIL: " #FUNC " (0.5) setting signgam"); \
> - result = 1; \
> - } \
> - signgam = 123; \
> - c = FUNC (b); \
> - if (signgam == -1) \
> - puts ("PASS: " #FUNC " (-0.5) setting signgam"); \
> - else \
> - { \
> - puts ("FAIL: " #FUNC " (-0.5) setting signgam"); \
> - result = 1; \
> - } \
> - } \
> - while (0)
> -
> -static int
> -do_test (void)
> -{
> - int result = 0;
> - RUN_TESTS (lgammaf, float);
> - RUN_TESTS (gammaf, float);
> - RUN_TESTS (lgamma, double);
> - RUN_TESTS (gamma, double);
> - RUN_TESTS (lgammal, long double);
> - RUN_TESTS (gammal, long double);
> - return result;
> -}
> -
> -#define TEST_FUNCTION do_test ()
> -#include "../test-skeleton.c"
>
Ok.
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH 2/4] Remove finite-math tests
2019-07-16 18:19 ` Adhemerval Zanella
@ 2019-10-18 18:14 ` Adhemerval Zanella
0 siblings, 0 replies; 3+ messages in thread
From: Adhemerval Zanella @ 2019-10-18 18:14 UTC (permalink / raw)
To: Wilco Dijkstra, libc-alpha; +Cc: nd
On 16/07/2019 15:19, Adhemerval Zanella wrote:
>
>
> On 26/06/2019 10:59, Wilco Dijkstra wrote:
>> Remove the finite-math tests from the testsuite - these are no longer
>> useful after removing math-finite.h header.
>>
>> Passes buildmanyglibc, build&test on x86_64 and AArch64.
>>
>> 2019-06-26 Wilco Dijkstra <wdijkstr@arm.com>
>>
>> * math/Makefile: Update to remove finite-math tests.
>> * math/gen-libm-test.py: Remove setting of NON_FINITE.
>> * math/libm-test-driver.c: Remove TEST_FINITE, flag_test_finite.
>> * math/libm-test-support.h: Remove flag_test_finite declartion.
>
> s/declartion/declaration
>
>> * math/libm-test-support.c: Remove NON_FINITE support.
>> * math/test-finite-macros.c: Remove file.
>> * math/test-math-finite.h: Likewise.
>> * math/test-math-inline.h: Remove TEST_FINITE.
>> * math/test-math-no-finite.h : Remove file.
>> * math/test-math-vector.h: Remove TEST_FINITE.
>> * math/test-signgam-finite.c: Remove test.
>> * math/test-signgam-finite-c11.c: Likewise.
>> * math/test-signgam-finite-c99.c: Likewise.
>
> LGTM with just one modification regarding TEST_EXCEPTIONS below.
>
> Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
I chatted with Wilco and I will commit this patch shortly with the
small modification I pointed out on his behalf.
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2019-10-18 18:14 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-06-26 13:59 [PATCH 2/4] Remove finite-math tests Wilco Dijkstra
2019-07-16 18:19 ` Adhemerval Zanella
2019-10-18 18:14 ` Adhemerval Zanella
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).