Go to file
Adhemerval Zanella 72a48e45bd math: Use erf from CORE-MATH
The current implementation precision shows the following accuracy, on
three rangeis ([-DBL_MIN, -4.2], [-4.2, 4.2], [4.2, DBL_MAX]) with
10e9 uniform randomly generated numbers for each range (first column
is the accuracy in ULP, with '0' being correctly rounded, second is the
number of samples with the corresponding precision):

* Range [-DBL_MIN, -4.2]
 * FE_TONEAREST
     0:      10000000000 100.00%
 * FE_UPWARD
     0:      10000000000 100.00%
 * FE_DOWNWARD
     0:      10000000000 100.00%
 * FE_TOWARDZERO
     0:      10000000000 100.00%

* Range [-4.2, 4.2]
 * FE_TONEAREST
     0:       9764404513  97.64%
     1:        235595487   2.36%
 * FE_UPWARD
     0:       9468013928  94.68%
     1:        531986072   5.32%
 * FE_DOWNWARD
     0:       9493787693  94.94%
     1:        506212307   5.06%
 * FE_TOWARDZERO
     0:       9585271351  95.85%
     1:        414728649   4.15%

* Range [4.2, DBL_MAX]
 * FE_TONEAREST
     0:      10000000000 100.00%
 * FE_UPWARD
     0:      10000000000 100.00%
 * FE_DOWNWARD
     0:      10000000000 100.00%
 * FE_TOWARDZERO
     0:      10000000000 100.00%

The CORE-MATH implementation is correctly rounded for any rounding mode.
The code was adapted to glibc style and to use the definition of
math_config.h (to handle errno, overflow, and underflow).

Benchtest on x64_64 (Ryzen 9 5900X, gcc 14.2.1), aarch64 (Neoverse-N1,
gcc 13.3.1), and powerpc (POWER10, gcc 13.2.1) shows:

reciprocal-throughput        master       patched   improvement
x86_64                      38.2754       78.0311      -103.87%
x86_64v2                    38.3325       75.7555       -97.63%
x86_64v3                    34.6604       28.3182        18.30%
aarch64                     23.1499       21.4307         7.43%
power10                     12.3051       9.3766         23.80%

Latency                      master       patched   improvement
x86_64                      84.3062      121.3580       -43.95%
x86_64v2                    84.1817      117.4250       -39.49%
x86_64v3                    81.0933       70.6458        12.88%
aarch64                      35.012       29.5012        15.74%
power10                     21.7205       18.4589        15.02%

For x86_64/x86_64-v2, most performance hit came from the fma call
through the ifunc mechanism.

Checked on x86_64-linux-gnu, aarch64-linux-gnu, and
powerpc64le-linux-gnu.

Reviewed-by: DJ Delorie <dj@redhat.com>
2025-10-27 09:34:04 -03:00
ChangeLog.old
advisories
argp
assert assert: Refactor assert/assert_perror 2025-09-23 10:29:24 -03:00
benchtests plot_strings.py: Replace np.complex with complex 2025-10-22 07:52:54 +08:00
bits Add once_flag, ONCE_FLAG_INIT and call_once to stdlib.h for C23 2025-10-01 15:15:15 +00:00
catgets catgets: Remove catgets/config.h 2025-10-21 09:24:19 -03:00
conform
csu
ctype
debug Implement C23 memset_explicit (bug 32378) 2025-10-01 15:14:09 +00:00
dirent
dlfcn
elf sprof: fix -Wformat warnings on 32-bit hosts 2025-10-22 09:48:44 -07:00
gmon gmon: Only used -fno-tree-loop-distribute-patterns if compiler supports it 2025-10-21 09:27:01 -03:00
gnulib
hesiod
htl
hurd Suppress -Wmaybe-uninitialized only for gcc 2025-10-21 09:24:05 -03:00
iconv Suppress -Wmaybe-uninitialized only for gcc 2025-10-21 09:24:05 -03:00
iconvdata iconvdata: Fix clang -Wstring-plus-int clang warning 2025-10-21 09:24:14 -03:00
include stdio: Only use __va_arg_pack if compiler supports it 2025-10-21 09:26:04 -03:00
inet
intl
io Tests: Create files with mode 0666, not 0777 (bug 33171) 2025-09-01 15:12:28 +02:00
libio libio: Add terminating NUL when the first character is EOF in getdelim [BZ #28038] 2025-10-09 17:38:01 -07:00
locale Suppress -Wmaybe-uninitialized only for gcc 2025-10-21 09:24:05 -03:00
localedata tests-mbwc/tst_funcs.h: Fix typo 2025-10-01 14:49:10 -03:00
login login: fix ut_line comparison logic 2025-08-29 14:49:19 -04:00
mach
malloc malloc: fix large tcache code to check for exact size match 2025-10-24 16:55:02 +00:00
manual Implement C23 memalignment 2025-10-17 16:56:59 +00:00
math math: Use erf from CORE-MATH 2025-10-27 09:34:04 -03:00
mathvec
misc misc: Fix clang -Wstring-plus-int warnings on syslog 2025-10-22 16:35:39 -03:00
nis Suppress -Wmaybe-uninitialized only for gcc 2025-10-21 09:24:05 -03:00
nptl nptl: Fix Wincompatible-pointer-types on clang 2025-10-21 09:24:06 -03:00
nptl_db Suppress -Wmaybe-uninitialized only for gcc 2025-10-21 09:24:05 -03:00
nscd atomic: Remove atomic_forced_read 2025-10-08 12:59:30 +00:00
nss Alphabetize nss/Makefile 2025-10-09 09:50:27 +02:00
po
posix posix: Fix memory leak a memory leak in glob. 2025-10-21 19:09:45 -07:00
resolv Suppress -Wmaybe-uninitialized only for gcc 2025-10-21 09:24:05 -03:00
resource
rt
scripts Use Linux 6.17 in build-many-glibcs.py 2025-10-01 15:15:43 +00:00
setjmp
signal
socket
soft-fp
stdio-common stdio: Fix -Wtautological-constant-out-of-range-compare on clang 2025-10-21 09:26:04 -03:00
stdlib Adjust stdint for clang-20 2025-10-20 11:33:54 -03:00
string Suppress -Wmaybe-uninitialized only for gcc 2025-10-21 09:24:05 -03:00
sunrpc sunrpc: Fix clang build 2025-10-21 09:27:05 -03:00
support support: Handle clang support/dtotimespec.c on dtotimespec 2025-10-21 09:26:04 -03:00
sysdeps math: Use erf from CORE-MATH 2025-10-27 09:34:04 -03:00
sysvipc
termios
time
timezone configure: Use -Wno-maybe-uninitialized iff compiler supports it 2025-10-21 09:24:03 -03:00
wcsmbs
wctype
.b4-config
.clang-format
.gitattributes
.gitignore
CONTRIBUTED-BY
COPYING
COPYING.LIB
INSTALL
LICENSES
MAINTAINERS
Makeconfig Build glibc with -ftrapping-math 2025-10-20 11:33:54 -03:00
Makefile
Makefile.help
Makefile.in Remove test-have-mamx-tile from Makefile.in 2025-08-23 15:58:00 -07:00
Makerules
NEWS Implement C23 memalignment 2025-10-17 16:56:59 +00:00
README
Rules
SECURITY.md
SHARED-FILES math: Use erf from CORE-MATH 2025-10-27 09:34:04 -03:00
abi-tags
aclocal.m4 x86: Set have-protected-data to no if unsupported 2025-08-22 17:55:32 -07:00
config.h.in hurd: implement RLIMIT_AS against Mach RPCs 2025-09-22 00:52:37 +02:00
config.make.in
configure Fix configure from ab22e5ec37 2025-10-22 17:23:12 -03:00
configure.ac gmon: Only used -fno-tree-loop-distribute-patterns if compiler supports it 2025-10-21 09:27:01 -03:00
extra-lib.mk
gen-locales.mk
libc-abis
libof-iterator.mk
o-iterator.mk
shlib-versions
test-skeleton.c
version.h

README

This directory contains the sources of the GNU C Library.
See the file "version.h" for what release version you have.

The GNU C Library is the standard system C library for all GNU systems,
and is an important part of what makes up a GNU system.  It provides the
system API for all programs written in C and C-compatible languages such
as C++ and Objective C; the runtime facilities of other programming
languages use the C library to access the underlying operating system.

In GNU/Linux systems, the C library works with the Linux kernel to
implement the operating system behavior seen by user applications.
In GNU/Hurd systems, it works with a microkernel and Hurd servers.

The GNU C Library implements much of the POSIX.1 functionality in the
GNU/Hurd system, using configurations i[4567]86-*-gnu and x86_64-gnu.

When working with Linux kernels, this version of the GNU C Library
requires Linux kernel version 3.2 or later.

Also note that the shared version of the libgcc_s library must be
installed for the pthread library to work correctly.

The GNU C Library supports these configurations for using Linux kernels:

	aarch64*-*-linux-gnu
	alpha*-*-linux-gnu
	arc*-*-linux-gnu
	arm-*-linux-gnueabi
	csky-*-linux-gnuabiv2
	hppa-*-linux-gnu
	i[4567]86-*-linux-gnu
	x86_64-*-linux-gnu	Can build either x86_64 or x32
	loongarch64-*-linux-gnu Hardware floating point, LE only.
	m68k-*-linux-gnu
	microblaze*-*-linux-gnu
	mips-*-linux-gnu
	mips64-*-linux-gnu
	or1k-*-linux-gnu
	powerpc-*-linux-gnu	Hardware or software floating point, BE only.
	powerpc64*-*-linux-gnu	Big-endian and little-endian.
	s390-*-linux-gnu
	s390x-*-linux-gnu
	riscv32-*-linux-gnu
	riscv64-*-linux-gnu
	sh[34]-*-linux-gnu
	sparc*-*-linux-gnu
	sparc64*-*-linux-gnu

If you are interested in doing a port, please contact the glibc
maintainers; see https://www.gnu.org/software/libc/ for more
information.

See the file INSTALL to find out how to configure, build, and install
the GNU C Library.  You might also consider reading the WWW pages for
the C library at https://www.gnu.org/software/libc/.

The GNU C Library is (almost) completely documented by the Texinfo manual
found in the `manual/' subdirectory.  The manual is still being updated
and contains some known errors and omissions; we regret that we do not
have the resources to work on the manual as much as we would like.  For
corrections to the manual, please file a bug in the `manual' component,
following the bug-reporting instructions below.  Please be sure to check
the manual in the current development sources to see if your problem has
already been corrected.

Please see https://www.gnu.org/software/libc/bugs.html for bug reporting
information.  We are now using the Bugzilla system to track all bug reports.
This web page gives detailed information on how to report bugs properly.

The GNU C Library is free software.  See the file COPYING.LIB for copying
conditions, and LICENSES for notices about a few contributions that require
these additional notices to be distributed.  License copyright years may be
listed using range notation, e.g., 1996-2015, indicating that every year in
the range, inclusive, is a copyrightable year that would otherwise be listed
individually.