[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Commits] r17687 - in /fsf/trunk/libc: ./ conform/ manual/ math/ sysdeps/i386/ sysdeps/i386/fpu/ sysdeps/ieee754/dbl-64/ sysdeps/unix/...
- To: commits@xxxxxxxxxx
- Subject: [Commits] r17687 - in /fsf/trunk/libc: ./ conform/ manual/ math/ sysdeps/i386/ sysdeps/i386/fpu/ sysdeps/ieee754/dbl-64/ sysdeps/unix/...
- From: eglibc@xxxxxxxxxx
- Date: Wed, 21 Mar 2012 19:51:00 -0000
Author: eglibc
Date: Wed Mar 21 19:50:58 2012
New Revision: 17687
Log:
Import glibc-mainline for 2012-03-21
Removed:
fsf/trunk/libc/NOTES
fsf/trunk/libc/sysdeps/i386/fpu/s_cexp.S
fsf/trunk/libc/sysdeps/i386/fpu/s_cexpf.S
fsf/trunk/libc/sysdeps/i386/fpu/s_cexpl.S
Modified:
fsf/trunk/libc/ChangeLog
fsf/trunk/libc/Makefile
fsf/trunk/libc/NEWS
fsf/trunk/libc/README
fsf/trunk/libc/aclocal.m4
fsf/trunk/libc/configure
fsf/trunk/libc/configure.in
fsf/trunk/libc/conform/conformtest.pl
fsf/trunk/libc/conform/run-conformtest.sh
fsf/trunk/libc/manual/creature.texi
fsf/trunk/libc/manual/intro.texi
fsf/trunk/libc/manual/lang.texi
fsf/trunk/libc/math/libm-test.inc
fsf/trunk/libc/math/w_exp2.c
fsf/trunk/libc/math/w_exp2f.c
fsf/trunk/libc/math/w_exp2l.c
fsf/trunk/libc/sysdeps/i386/configure
fsf/trunk/libc/sysdeps/i386/configure.in
fsf/trunk/libc/sysdeps/i386/fpu/e_pow.S
fsf/trunk/libc/sysdeps/i386/fpu/e_powf.S
fsf/trunk/libc/sysdeps/i386/fpu/e_powl.S
fsf/trunk/libc/sysdeps/i386/fpu/libm-test-ulps
fsf/trunk/libc/sysdeps/ieee754/dbl-64/e_pow.c
fsf/trunk/libc/sysdeps/unix/sysv/linux/configure
fsf/trunk/libc/sysdeps/unix/sysv/linux/configure.in
fsf/trunk/libc/sysdeps/x86_64/fpu/e_powl.S
fsf/trunk/libc/sysdeps/x86_64/fpu/libm-test-ulps
fsf/trunk/libc/timezone/Makefile
Modified: fsf/trunk/libc/ChangeLog
==============================================================================
--- fsf/trunk/libc/ChangeLog (original)
+++ fsf/trunk/libc/ChangeLog Wed Mar 21 19:50:58 2012
@@ -1,3 +1,94 @@
+2012-03-21 H.J. Lu <hongjiu.lu@xxxxxxxxx>
+
+ * sysdeps/unix/sysv/linux/configure.in: Check x86_64* instead
+ of x86_64 when setting libc_cv_slibdir, libdir and
+ libc_cv_localedir.
+ * sysdeps/unix/sysv/linux/configure: Regenerated.
+
+2012-03-21 Joseph Myers <joseph@xxxxxxxxxxxxxxxx>
+
+ * manual/lang.texi (Old Varargs): Remove section.
+ (How Variadic): Update menu.
+ (va_start): Do not mention varargs.h.
+
+2012-03-21 Thomas Schwinge <thomas@xxxxxxxxxxxxxxxx>
+ Joseph Myers <joseph@xxxxxxxxxxxxxxxx>
+
+ * configure.in (libc_cv_ssp): Use LIBC_TRY_CC_OPTION instead of a
+ link test.
+ * configure: Regenerated.
+
+2012-03-21 Thomas Schwinge <thomas@xxxxxxxxxxxxxxxx>
+
+ * conform/conformtest.pl: Handle --tmpdir argument, defaulting to /tmp.
+ * conform/run-conformtest.sh: Pass --tmpdir argument when invoking
+ conformtest.pl
+
+2012-03-21 Joseph Myers <joseph@xxxxxxxxxxxxxxxx>
+
+ * NOTES: Remove.
+ * Makefile (files-for-dist): Remove NOTES.
+ (NOTES): Remove rule.
+ * README: Don't refer to NOTES.
+ * manual/creature.texi: Don't include macros.texi.
+ * manual/intro.texi (creature.texi): Remove comment referring to
+ NOTES.
+
+ * aclocal.m4 (LIBC_TRY_CC_OPTION): New macro.
+ * configure.in (libc_cv_cc_submachine): Use LIBC_TRY_CC_OPTION.
+ * configure: Regenerated.
+ * sysdeps/i386/configure.in (libc_cv_cc_sse4): Use
+ LIBC_TRY_CC_OPTION.
+ (libc_cv_as_i686): Likewise.
+ (libc_cv_cc_avx): Likewise.
+ (libc_cv_cc_sse2avx): Likewise.
+ (libc_cv_cc_fma4): Likewise.
+ (libc_cv_cc_novzeroupper): Likewise.
+ * sysdeps/i386/configure: Regenerated.
+
+ [BZ #13883]
+ * sysdeps/i386/fpu/s_cexp.S: Remove.
+ * sysdeps/i386/fpu/s_cexpf.S: Likewise.
+ * sysdeps/i386/fpu/s_cexpl.S: Likewise.
+ * math/libm-test.inc (cexp_test): Add more tests.
+ * sysdeps/i386/fpu/libm-test-ulps: Update.
+ * sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
+
+2012-03-21 Allan McRae <allan@xxxxxxxxxxxxx>
+
+ * timezone/Makefile: Do not install iso3166.tab and zone.tab
+
+2012-03-21 Joseph Myers <joseph@xxxxxxxxxxxxxxxx>
+
+ [BZ #13871]
+ * math/w_exp2.c: Do not include <float.h>.
+ (o_threshold, u_threshold): Remove.
+ (__exp2): Calculate result before checking finiteness and calling
+ __kernel_standard.
+ * math/w_exp2f.c: Likewise.
+ * math/w_exp2l.c: Likewise.
+ * math/libm-test.inc (exp2_test): Require overflow exception for
+ 1e6 input.
+
+ [BZ #3866]
+ * sysdeps/i386/fpu/e_pow.S (__ieee754_pow): Test for y outside the
+ range of signed 64-bit integers before using fistpll. Remove
+ checks for whether integers fit in mantissa bits.
+ * sysdeps/i386/fpu/e_powf.S (__ieee754_powf): Test for y outside
+ the range of signed 32-bit integers before using fistpl. Remove
+ checks for whether integers fit in mantissa bits.
+ * sysdeps/i386/fpu/e_powl.S (p64): New object.
+ (__ieee754_powl): Test for y outside the range of signed 64-bit
+ integers before using fistpll. Reduce 64-bit values to 63-bit
+ ones as needed.
+ * sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Ensure
+ divide-by-zero is raised for zero to large negative powers.
+ * sysdeps/x86_64/fpu/e_powl.S (p64): New object.
+ (__ieee754_powl): Test for y outside the range of signed 64-bit
+ integers before using fistpll. Reduce 64-bit values to 63-bit
+ ones as needed.
+ * math/libm-test.inc (pow_test): Add more tests.
+
2012-03-20 H.J. Lu <hongjiu.lu@xxxxxxxxx>
* debug/backtracesymsfd.c: Include <_itoa.h> instead of
Modified: fsf/trunk/libc/Makefile
==============================================================================
--- fsf/trunk/libc/Makefile (original)
+++ fsf/trunk/libc/Makefile Wed Mar 21 19:50:58 2012
@@ -1,5 +1,4 @@
-# Copyright (C) 1991-2002,2003,2004,2005,2006,2008,2009,2011
-# Free Software Foundation, Inc.
+# Copyright (C) 1991-2012 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
@@ -364,7 +363,7 @@
generated := $(generated) stubs.h
-files-for-dist := README FAQ INSTALL NOTES configure ChangeLog NEWS
+files-for-dist := README FAQ INSTALL configure ChangeLog NEWS
# Regenerate stuff, then error if these things are not committed yet.
dist-prepare: $(files-for-dist)
@@ -399,7 +398,6 @@
-chmod a-w $@
endef
INSTALL: manual/install.texi manual/macros.texi; $(format-me)
-NOTES: manual/creature.texi manual/macros.texi; $(format-me)
manual/dir-add.texi manual/dir-add.info: FORCE
$(MAKE) $(PARALLELMFLAGS) -C $(@D) $(@F)
FAQ: scripts/gen-FAQ.pl FAQ.in
Modified: fsf/trunk/libc/NEWS
==============================================================================
--- fsf/trunk/libc/NEWS (original)
+++ fsf/trunk/libc/NEWS Wed Mar 21 19:50:58 2012
@@ -16,7 +16,7 @@
13526, 13527, 13528, 13529, 13530, 13531, 13532, 13533, 13547, 13551,
13552, 13553, 13555, 13559, 13566, 13583, 13618, 13637, 13656, 13658,
13673, 13695, 13704, 13706, 13726, 13738, 13786, 13792, 13806, 13840,
- 13841, 13844, 13846, 13851, 13852, 13854
+ 13841, 13844, 13846, 13851, 13852, 13854, 13871, 13883
* ISO C11 support:
Removed: fsf/trunk/libc/NOTES
==============================================================================
--- fsf/trunk/libc/NOTES (original)
+++ fsf/trunk/libc/NOTES (removed)
@@ -1,223 +1,0 @@
-Feature Test Macros
--------------------
-
-The exact set of features available when you compile a source file is
-controlled by which "feature test macros" you define.
-
- If you compile your programs using `gcc -ansi', you get only the
-ISO C library features, unless you explicitly request additional
-features by defining one or more of the feature macros. *Note GNU CC
-Command Options: (gcc.info)Invoking GCC, for more information about GCC
-options.
-
- You should define these macros by using `#define' preprocessor
-directives at the top of your source code files. These directives
-_must_ come before any `#include' of a system header file. It is best
-to make them the very first thing in the file, preceded only by
-comments. You could also use the `-D' option to GCC, but it's better
-if you make the source files indicate their own meaning in a
-self-contained way.
-
- This system exists to allow the library to conform to multiple
-standards. Although the different standards are often described as
-supersets of each other, they are usually incompatible because larger
-standards require functions with names that smaller ones reserve to the
-user program. This is not mere pedantry -- it has been a problem in
-practice. For instance, some non-GNU programs define functions named
-`getline' that have nothing to do with this library's `getline'. They
-would not be compilable if all features were enabled indiscriminately.
-
- This should not be used to verify that a program conforms to a
-limited standard. It is insufficient for this purpose, as it will not
-protect you from including header files outside the standard, or
-relying on semantics undefined within the standard.
-
- -- Macro: _POSIX_SOURCE
- If you define this macro, then the functionality from the POSIX.1
- standard (IEEE Standard 1003.1) is available, as well as all of the
- ISO C facilities.
-
- The state of `_POSIX_SOURCE' is irrelevant if you define the macro
- `_POSIX_C_SOURCE' to a positive integer.
-
- -- Macro: _POSIX_C_SOURCE
- Define this macro to a positive integer to control which POSIX
- functionality is made available. The greater the value of this
- macro, the more functionality is made available.
-
- If you define this macro to a value greater than or equal to `1',
- then the functionality from the 1990 edition of the POSIX.1
- standard (IEEE Standard 1003.1-1990) is made available.
-
- If you define this macro to a value greater than or equal to `2',
- then the functionality from the 1992 edition of the POSIX.2
- standard (IEEE Standard 1003.2-1992) is made available.
-
- If you define this macro to a value greater than or equal to
- `199309L', then the functionality from the 1993 edition of the
- POSIX.1b standard (IEEE Standard 1003.1b-1993) is made available.
-
- Greater values for `_POSIX_C_SOURCE' will enable future extensions.
- The POSIX standards process will define these values as necessary,
- and the GNU C Library should support them some time after they
- become standardized. The 1996 edition of POSIX.1 (ISO/IEC 9945-1:
- 1996) states that if you define `_POSIX_C_SOURCE' to a value
- greater than or equal to `199506L', then the functionality from
- the 1996 edition is made available.
-
- -- Macro: _BSD_SOURCE
- If you define this macro, functionality derived from 4.3 BSD Unix
- is included as well as the ISO C, POSIX.1, and POSIX.2 material.
-
- Some of the features derived from 4.3 BSD Unix conflict with the
- corresponding features specified by the POSIX.1 standard. If this
- macro is defined, the 4.3 BSD definitions take precedence over the
- POSIX definitions.
-
- Due to the nature of some of the conflicts between 4.3 BSD and
- POSIX.1, you need to use a special "BSD compatibility library"
- when linking programs compiled for BSD compatibility. This is
- because some functions must be defined in two different ways, one
- of them in the normal C library, and one of them in the
- compatibility library. If your program defines `_BSD_SOURCE', you
- must give the option `-lbsd-compat' to the compiler or linker when
- linking the program, to tell it to find functions in this special
- compatibility library before looking for them in the normal C
- library.
-
- -- Macro: _SVID_SOURCE
- If you define this macro, functionality derived from SVID is
- included as well as the ISO C, POSIX.1, POSIX.2, and X/Open
- material.
-
- -- Macro: _XOPEN_SOURCE
- -- Macro: _XOPEN_SOURCE_EXTENDED
- If you define this macro, functionality described in the X/Open
- Portability Guide is included. This is a superset of the POSIX.1
- and POSIX.2 functionality and in fact `_POSIX_SOURCE' and
- `_POSIX_C_SOURCE' are automatically defined.
-
- As the unification of all Unices, functionality only available in
- BSD and SVID is also included.
-
- If the macro `_XOPEN_SOURCE_EXTENDED' is also defined, even more
- functionality is available. The extra functions will make all
- functions available which are necessary for the X/Open Unix brand.
-
- If the macro `_XOPEN_SOURCE' has the value 500 this includes all
- functionality described so far plus some new definitions from the
- Single Unix Specification, version 2.
-
- -- Macro: _LARGEFILE_SOURCE
- If this macro is defined some extra functions are available which
- rectify a few shortcomings in all previous standards.
- Specifically, the functions `fseeko' and `ftello' are available.
- Without these functions the difference between the ISO C interface
- (`fseek', `ftell') and the low-level POSIX interface (`lseek')
- would lead to problems.
-
- This macro was introduced as part of the Large File Support
- extension (LFS).
-
- -- Macro: _LARGEFILE64_SOURCE
- If you define this macro an additional set of functions is made
- available which enables 32 bit systems to use files of sizes beyond
- the usual limit of 2GB. This interface is not available if the
- system does not support files that large. On systems where the
- natural file size limit is greater than 2GB (i.e., on 64 bit
- systems) the new functions are identical to the replaced functions.
-
- The new functionality is made available by a new set of types and
- functions which replace the existing ones. The names of these new
- objects contain `64' to indicate the intention, e.g., `off_t' vs.
- `off64_t' and `fseeko' vs. `fseeko64'.
-
- This macro was introduced as part of the Large File Support
- extension (LFS). It is a transition interface for the period when
- 64 bit offsets are not generally used (see `_FILE_OFFSET_BITS').
-
- -- Macro: _FILE_OFFSET_BITS
- This macro determines which file system interface shall be used,
- one replacing the other. Whereas `_LARGEFILE64_SOURCE' makes the
- 64 bit interface available as an additional interface,
- `_FILE_OFFSET_BITS' allows the 64 bit interface to replace the old
- interface.
-
- If `_FILE_OFFSET_BITS' is undefined, or if it is defined to the
- value `32', nothing changes. The 32 bit interface is used and
- types like `off_t' have a size of 32 bits on 32 bit systems.
-
- If the macro is defined to the value `64', the large file interface
- replaces the old interface. I.e., the functions are not made
- available under different names (as they are with
- `_LARGEFILE64_SOURCE'). Instead the old function names now
- reference the new functions, e.g., a call to `fseeko' now indeed
- calls `fseeko64'.
-
- This macro should only be selected if the system provides
- mechanisms for handling large files. On 64 bit systems this macro
- has no effect since the `*64' functions are identical to the
- normal functions.
-
- This macro was introduced as part of the Large File Support
- extension (LFS).
-
- -- Macro: _ISOC99_SOURCE
- Until the revised ISO C standard is widely adopted the new features
- are not automatically enabled. The GNU C Library nevertheless has
- a complete implementation of the new standard and to enable the
- new features the macro `_ISOC99_SOURCE' should be defined.
-
- -- Macro: _GNU_SOURCE
- If you define this macro, everything is included: ISO C89,
- ISO C99, POSIX.1, POSIX.2, BSD, SVID, X/Open, LFS, and GNU
- extensions. In the cases where POSIX.1 conflicts with BSD, the
- POSIX definitions take precedence.
-
- If you want to get the full effect of `_GNU_SOURCE' but make the
- BSD definitions take precedence over the POSIX definitions, use
- this sequence of definitions:
-
- #define _GNU_SOURCE
- #define _BSD_SOURCE
- #define _SVID_SOURCE
-
- Note that if you do this, you must link your program with the BSD
- compatibility library by passing the `-lbsd-compat' option to the
- compiler or linker. *NB:* If you forget to do this, you may get
- very strange errors at run time.
-
- -- Macro: _REENTRANT
- -- Macro: _THREAD_SAFE
- If you define one of these macros, reentrant versions of several
- functions get declared. Some of the functions are specified in
- POSIX.1c but many others are only available on a few other systems
- or are unique to the GNU C Library. The problem is the delay in
- the standardization of the thread safe C library interface.
-
- Unlike on some other systems, no special version of the C library
- must be used for linking. There is only one version but while
- compiling this it must have been specified to compile as thread
- safe.
-
- We recommend you use `_GNU_SOURCE' in new programs. If you don't
-specify the `-ansi' option to GCC and don't define any of these macros
-explicitly, the effect is the same as defining `_POSIX_C_SOURCE' to 2
-and `_POSIX_SOURCE', `_SVID_SOURCE', and `_BSD_SOURCE' to 1.
-
- When you define a feature test macro to request a larger class of
-features, it is harmless to define in addition a feature test macro for
-a subset of those features. For example, if you define
-`_POSIX_C_SOURCE', then defining `_POSIX_SOURCE' as well has no effect.
-Likewise, if you define `_GNU_SOURCE', then defining either
-`_POSIX_SOURCE' or `_POSIX_C_SOURCE' or `_SVID_SOURCE' as well has no
-effect.
-
- Note, however, that the features of `_BSD_SOURCE' are not a subset of
-any of the other feature test macros supported. This is because it
-defines BSD features that take precedence over the POSIX features that
-are requested by the other macros. For this reason, defining
-`_BSD_SOURCE' in addition to the other feature test macros does have an
-effect: it causes the BSD features to take priority over the conflicting
-POSIX features.
-
Modified: fsf/trunk/libc/README
==============================================================================
--- fsf/trunk/libc/README (original)
+++ fsf/trunk/libc/README Wed Mar 21 19:50:58 2012
@@ -84,10 +84,6 @@
the manual in the current development sources to see if your problem has
already been corrected.
-The file NOTES contains a description of the feature-test macros used
-in the GNU C library, explaining how you can tell the library what
-facilities you want it to make available.
-
Please see http://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.
Modified: fsf/trunk/libc/aclocal.m4
==============================================================================
--- fsf/trunk/libc/aclocal.m4 (original)
+++ fsf/trunk/libc/aclocal.m4 Wed Mar 21 19:50:58 2012
@@ -128,6 +128,12 @@
[$2], [$3])
rm -f conftest*])
+dnl Test a compiler option or options with an empty input file.
+dnl LIBC_TRY_CC_OPTION([options], [action-if-true], [action-if-false])
+AC_DEFUN([LIBC_TRY_CC_OPTION],
+[AS_IF([AC_TRY_COMMAND([${CC-cc} $1 -xc /dev/null -S -o /dev/null])],
+ [$2], [$3])])
+
dnl Find and source sysdeps/*/preconfigure.
dnl LIBC_PRECONFIGURE([$srcdir], [for])
AC_DEFUN([LIBC_PRECONFIGURE], [dnl
Modified: fsf/trunk/libc/configure
==============================================================================
--- fsf/trunk/libc/configure (original)
+++ fsf/trunk/libc/configure Wed Mar 21 19:50:58 2012
@@ -6643,23 +6643,17 @@
if ${libc_cv_ssp+:} false; then :
$as_echo_n "(cached) " >&6
else
- cat > conftest.c <<EOF
-int foo;
-main () { return 0;}
-EOF
-if { ac_try='${CC-cc} $CFLAGS $CPPFLAGS $LDFLAGS -Werror -fstack-protector
- -o conftest conftest.c 1>&5'
+ if { ac_try='${CC-cc} $CFLAGS $CPPFLAGS -Werror -fstack-protector -xc /dev/null -S -o /dev/null'
{ { eval echo "\"\$as_me\":${as_lineno-$LINENO}: \"$ac_try\""; } >&5
(eval $ac_try) 2>&5
ac_status=$?
$as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5
- test $ac_status = 0; }; }
-then
+ test $ac_status = 0; }; }; then :
libc_cv_ssp=yes
else
libc_cv_ssp=no
fi
-rm -f conftest*
+
fi
{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $libc_cv_ssp" >&5
$as_echo "$libc_cv_ssp" >&6; }
@@ -7054,10 +7048,11 @@
(eval $ac_try) 2>&5
ac_status=$?
$as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5
- test $ac_status = 0; }; }; then
+ test $ac_status = 0; }; }; then :
+
libc_cv_cc_submachine="$opt"
break
- fi
+fi
done
fi
{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $libc_cv_cc_submachine" >&5
Modified: fsf/trunk/libc/configure.in
==============================================================================
--- fsf/trunk/libc/configure.in (original)
+++ fsf/trunk/libc/configure.in Wed Mar 21 19:50:58 2012
@@ -1753,18 +1753,10 @@
AC_SUBST(fno_unit_at_a_time)
AC_CACHE_CHECK(for -fstack-protector, libc_cv_ssp, [dnl
-cat > conftest.c <<EOF
-int foo;
-main () { return 0;}
-EOF
-if AC_TRY_COMMAND([${CC-cc} $CFLAGS $CPPFLAGS $LDFLAGS -Werror -fstack-protector
- -o conftest conftest.c 1>&AS_MESSAGE_LOG_FD])
-then
- libc_cv_ssp=yes
-else
- libc_cv_ssp=no
-fi
-rm -f conftest*])
+LIBC_TRY_CC_OPTION([$CFLAGS $CPPFLAGS -Werror -fstack-protector],
+ [libc_cv_ssp=yes],
+ [libc_cv_ssp=no])
+])
AC_SUBST(libc_cv_ssp)
AC_CACHE_CHECK(for -fgnu89-inline, libc_cv_gnu89_inline, [dnl
@@ -2005,10 +1997,9 @@
libc_cv_cc_submachine, [dnl
libc_cv_cc_submachine=no
for opt in "-march=$submachine" "-mcpu=$submachine"; do
- if AC_TRY_COMMAND([${CC-cc} $opt -xc /dev/null -S -o /dev/null]); then
+ LIBC_TRY_CC_OPTION([$opt], [
libc_cv_cc_submachine="$opt"
- break
- fi
+ break], [])
done])
if test "x$libc_cv_cc_submachine" = xno; then
AC_MSG_ERROR([${CC-cc} does not support $submachine])
Modified: fsf/trunk/libc/conform/conformtest.pl
==============================================================================
--- fsf/trunk/libc/conform/conformtest.pl (original)
+++ fsf/trunk/libc/conform/conformtest.pl Wed Mar 21 19:50:58 2012
@@ -3,11 +3,11 @@
use Getopt::Long;
use POSIX;
+$standard = "XOPEN2K8";
$CC = "gcc";
-
-$standard="XOPEN2K8";
+$tmpdir = "/tmp";
GetOptions ('headers=s' => \@headers, 'standard=s' => \$standard,
- 'flags=s' => \$flags, 'cc=s' => \$CC);
+ 'flags=s' => \$flags, 'cc=s' => \$CC, 'tmpdir=s' => \$tmpdir);
@headers = split(/,/,join(',',@headers));
# List of the headers we are testing.
@@ -83,11 +83,6 @@
while ($#knownproblems >= 0) {
$isknown{pop (@knownproblems)} = 1;
}
-
-$uid = getuid();
-($pwname,$pwpasswd,$pwuid,$pwgid,
- $pwquota,$pwcomment,$pwgcos,$pwdir,$pwshell,$pwexpire) = getpwuid($uid);
-$tmpdir = "$pwdir";
$verbose = 1;
Modified: fsf/trunk/libc/conform/run-conformtest.sh
==============================================================================
--- fsf/trunk/libc/conform/run-conformtest.sh (original)
+++ fsf/trunk/libc/conform/run-conformtest.sh Wed Mar 21 19:50:58 2012
@@ -4,6 +4,9 @@
perl="$2"
cc="$3"
includes="$4"
+
+scratch=${objpfx}scratch
+mkdir -p "$scratch"
standards=("ISO" "ISO99" "ISO11" "POSIX" "XPG3" "XPG4" "UNIX98"
"XOPEN2K" "XOPEN2K8" "POSIX2008")
@@ -13,7 +16,7 @@
for s in ${standards[*]}; do
echo -n $s...
e=0
- if ! $perl conformtest.pl --cc="$cc" --flags="$includes" --standard=$s > ${objpfx}conform-$s.out; then
+ if ! $perl conformtest.pl --tmpdir="$scratch" --cc="$cc" --flags="$includes" --standard=$s > ${objpfx}conform-$s.out; then
e=1
fi
printf "\n%s\n" $s >> ${objpfx}run-conformtest.out
Modified: fsf/trunk/libc/manual/creature.texi
==============================================================================
--- fsf/trunk/libc/manual/creature.texi (original)
+++ fsf/trunk/libc/manual/creature.texi Wed Mar 21 19:50:58 2012
@@ -1,6 +1,5 @@
@node Feature Test Macros
@subsection Feature Test Macros
-@include macros.texi
@cindex feature test macros
The exact set of features available when you compile a source file
Modified: fsf/trunk/libc/manual/intro.texi
==============================================================================
--- fsf/trunk/libc/manual/intro.texi (original)
+++ fsf/trunk/libc/manual/intro.texi Wed Mar 21 19:50:58 2012
@@ -548,7 +548,6 @@
@end itemize
@comment Include the section on Creature Nest Macros.
-@comment It is in a separate file so it can be formatted into ../NOTES.
@include creature.texi
@node Roadmap to the Manual, , Using the Library, Introduction
Modified: fsf/trunk/libc/manual/lang.texi
==============================================================================
--- fsf/trunk/libc/manual/lang.texi (original)
+++ fsf/trunk/libc/manual/lang.texi Wed Mar 21 19:50:58 2012
@@ -229,7 +229,6 @@
variable arguments functions.
* Argument Macros:: Detailed specification of the macros
for accessing variable arguments.
-* Old Varargs:: The pre-ISO way of defining variadic functions.
@end menu
@node Variadic Prototypes
@@ -427,9 +426,6 @@
This macro initializes the argument pointer variable @var{ap} to point
to the first of the optional arguments of the current function;
@var{last-required} must be the last required argument to the function.
-
-@xref{Old Varargs}, for an alternate definition of @code{va_start}
-found in the header file @file{varargs.h}.
@end deftypefn
@comment stdarg.h
@@ -511,68 +507,6 @@
@include add.c.texi
@end smallexample
-@node Old Varargs
-@subsubsection Old-Style Variadic Functions
-
-@pindex varargs.h
-Before @w{ISO C}, programmers used a slightly different facility for
-writing variadic functions. The GNU C compiler still supports it;
-currently, it is more portable than the @w{ISO C} facility, since support
-for @w{ISO C} is still not universal. The header file which defines the
-old-fashioned variadic facility is called @file{varargs.h}.
-
-Using @file{varargs.h} is almost the same as using @file{stdarg.h}.
-There is no difference in how you call a variadic function;
-see @ref{Calling Variadics}. The only difference is in how you define
-them. First of all, you must use old-style non-prototype syntax, like
-this:
-
-@smallexample
-tree
-build (va_alist)
- va_dcl
-@{
-@end smallexample
-
-Secondly, you must give @code{va_start} only one argument, like this:
-
-@smallexample
- va_list p;
- va_start (p);
-@end smallexample
-
-These are the special macros used for defining old-style variadic
-functions:
-
-@comment varargs.h
-@comment Unix
-@deffn Macro va_alist
-This macro stands for the argument name list required in a variadic
-function.
-@end deffn
-
-@comment varargs.h
-@comment Unix
-@deffn Macro va_dcl
-This macro declares the implicit argument or arguments for a variadic
-function.
-@end deffn
-
-@comment varargs.h
-@comment Unix
-@deftypefn {Macro} void va_start (va_list @var{ap})
-This macro, as defined in @file{varargs.h}, initializes the argument
-pointer variable @var{ap} to point to the first argument of the current
-function.
-@end deftypefn
-
-The other argument macros, @code{va_arg} and @code{va_end}, are the same
-in @file{varargs.h} as in @file{stdarg.h}; see @ref{Argument Macros}, for
-details.
-
-It does not work to include both @file{varargs.h} and @file{stdarg.h} in
-the same compilation; they define @code{va_start} in conflicting ways.
-
@node Null Pointer Constant
@section Null Pointer Constant
@cindex null pointer constant
Modified: fsf/trunk/libc/math/libm-test.inc
==============================================================================
--- fsf/trunk/libc/math/libm-test.inc (original)
+++ fsf/trunk/libc/math/libm-test.inc Wed Mar 21 19:50:58 2012
@@ -1894,6 +1894,21 @@
TEST_c_c (cexp, 0.75L, 1.25L, 0.667537446429131586942201977015932112L, 2.00900045494094876258347228145863909L);
TEST_c_c (cexp, -2.0, -3.0, -0.13398091492954261346140525546115575L, -0.019098516261135196432576240858800925L);
+ TEST_c_c (cexp, 0, 0x1p65, 0.99888622066058013610642172179340364209972L, -0.047183876212354673805106149805700013943218L);
+ TEST_c_c (cexp, 0, -0x1p65, 0.99888622066058013610642172179340364209972L, 0.047183876212354673805106149805700013943218L);
+ TEST_c_c (cexp, 50, 0x1p127, 4.053997150228616856622417636046265337193e21L, 3.232070315463388524466674772633810238819e21L);
+
+#ifndef TEST_FLOAT
+ TEST_c_c (cexp, 0, 1e22, 0.5232147853951389454975944733847094921409L, -0.8522008497671888017727058937530293682618L);
+ TEST_c_c (cexp, 0, 0x1p1023, -0.826369834614147994500785680811743734805L, 0.5631277798508840134529434079444683477104L);
+ TEST_c_c (cexp, 500, 0x1p1023, -1.159886268932754433233243794561351783426e217L, 7.904017694554466595359379965081774849708e216L);
+#endif
+
+#if defined TEST_LDOUBLE && LDBL_MAX_EXP >= 16384
+ TEST_c_c (cexp, 0, 0x1p16383L, 0.9210843909921906206874509522505756251609L, 0.3893629985894208126948115852610595405563L);
+ TEST_c_c (cexp, -10000, 0x1p16383L, 1.045876464564882298442774542991176546722e-4343L, 4.421154026488516836023811173959413420548e-4344L);
+#endif
+
END (cexp, complex);
}
@@ -3108,8 +3123,7 @@
TEST_f_f (exp2, 10, 1024);
TEST_f_f (exp2, -1, 0.5);
- /* Bug 13871: OVERFLOW exception may be missing. */
- TEST_f_f (exp2, 1e6, plus_infty, OVERFLOW_EXCEPTION_OK);
+ TEST_f_f (exp2, 1e6, plus_infty, OVERFLOW_EXCEPTION);
TEST_f_f (exp2, -1e6, 0);
TEST_f_f (exp2, 0.75L, 1.68179283050742908606225095246642979L);
@@ -5431,11 +5445,75 @@
TEST_ff_f (pow, 0, -11, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
check_int ("errno for pow(0,-odd) == ERANGE", errno, ERANGE, 0, 0, 0);
errno = 0;
+ TEST_ff_f (pow, 0, -0xffffff, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(0,-odd) == ERANGE", errno, ERANGE, 0, 0, 0);
+#ifndef TEST_FLOAT
+ errno = 0;
+ TEST_ff_f (pow, 0, -0x1.fffffffffffffp+52L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(0,-odd) == ERANGE", errno, ERANGE, 0, 0, 0);
+#endif
+#ifdef TEST_LDOUBLE
+# if LDBL_MANT_DIG >= 64
+ errno = 0;
+ TEST_ff_f (pow, 0, -0x1.fffffffffffffffep+63L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(0,-odd) == ERANGE", errno, ERANGE, 0, 0, 0);
+# endif
+# if LDBL_MANT_DIG >= 106
+ errno = 0;
+ TEST_ff_f (pow, 0, -0x1.ffffffffffffffffffffffffff8p+105L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(0,-odd) == ERANGE", errno, ERANGE, 0, 0, 0);
+# endif
+# if LDBL_MANT_DIG >= 113
+ errno = 0;
+ TEST_ff_f (pow, 0, -0x1.ffffffffffffffffffffffffffffp+112L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(0,-odd) == ERANGE", errno, ERANGE, 0, 0, 0);
+# endif
+#endif
TEST_ff_f (pow, minus_zero, -1, minus_infty, DIVIDE_BY_ZERO_EXCEPTION);
check_int ("errno for pow(-0,-odd) == ERANGE", errno, ERANGE, 0, 0, 0);
errno = 0;
TEST_ff_f (pow, minus_zero, -11L, minus_infty, DIVIDE_BY_ZERO_EXCEPTION);
check_int ("errno for pow(-0,-odd) == ERANGE", errno, ERANGE, 0, 0, 0);
+ errno = 0;
+ TEST_ff_f (pow, minus_zero, -0xffffff, minus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(-0,-odd) == ERANGE", errno, ERANGE, 0, 0, 0);
+ errno = 0;
+ TEST_ff_f (pow, minus_zero, -0x1fffffe, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(-0,-even) == ERANGE", errno, ERANGE, 0, 0, 0);
+#ifndef TEST_FLOAT
+ errno = 0;
+ TEST_ff_f (pow, minus_zero, -0x1.fffffffffffffp+52L, minus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(-0,-odd) == ERANGE", errno, ERANGE, 0, 0, 0);
+ errno = 0;
+ TEST_ff_f (pow, minus_zero, -0x1.fffffffffffffp+53L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(-0,-even) == ERANGE", errno, ERANGE, 0, 0, 0);
+#endif
+#ifdef TEST_LDOUBLE
+# if LDBL_MANT_DIG >= 64
+ errno = 0;
+ TEST_ff_f (pow, minus_zero, -0x1.fffffffffffffffep+63L, minus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(-0,-odd) == ERANGE", errno, ERANGE, 0, 0, 0);
+ errno = 0;
+ TEST_ff_f (pow, minus_zero, -0x1.fffffffffffffffep+64L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(-0,-even) == ERANGE", errno, ERANGE, 0, 0, 0);
+# endif
+# if LDBL_MANT_DIG >= 106
+ errno = 0;
+ TEST_ff_f (pow, minus_zero, -0x1.ffffffffffffffffffffffffff8p+105L, minus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(-0,-odd) == ERANGE", errno, ERANGE, 0, 0, 0);
+ errno = 0;
+ TEST_ff_f (pow, minus_zero, -0x1.ffffffffffffffffffffffffff8p+106L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(-0,-even) == ERANGE", errno, ERANGE, 0, 0, 0);
+# endif
+# if LDBL_MANT_DIG >= 113
+ errno = 0;
+ TEST_ff_f (pow, minus_zero, -0x1.ffffffffffffffffffffffffffffp+112L, minus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(-0,-odd) == ERANGE", errno, ERANGE, 0, 0, 0);
+ errno = 0;
+ TEST_ff_f (pow, minus_zero, -0x1.ffffffffffffffffffffffffffffp+113L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(-0,-even) == ERANGE", errno, ERANGE, 0, 0, 0);
+# endif
+#endif
errno = 0;
TEST_ff_f (pow, 0, -2, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
@@ -5444,10 +5522,30 @@
TEST_ff_f (pow, 0, -11.1L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
check_int ("errno for pow(0,-num) == ERANGE", errno, ERANGE, 0, 0, 0);
errno = 0;
+ TEST_ff_f (pow, 0, -0x1p24, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(0,-num) == ERANGE", errno, ERANGE, 0, 0, 0);
+ errno = 0;
+ TEST_ff_f (pow, 0, -0x1p127, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(0,-num) == ERANGE", errno, ERANGE, 0, 0, 0);
+ errno = 0;
+ /* Bug 13879: spurious OVERFLOW exception may be present. */
+ TEST_ff_f (pow, 0, -max_value, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|OVERFLOW_EXCEPTION_OK);
+ check_int ("errno for pow(0,-num) == ERANGE", errno, ERANGE, 0, 0, 0);
+ errno = 0;
TEST_ff_f (pow, minus_zero, -2, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
check_int ("errno for pow(-0,-even) == ERANGE", errno, ERANGE, 0, 0, 0);
errno = 0;
TEST_ff_f (pow, minus_zero, -11.1L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(-0,-num) == ERANGE", errno, ERANGE, 0, 0, 0);
+ errno = 0;
+ TEST_ff_f (pow, minus_zero, -0x1p24, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(-0,-num) == ERANGE", errno, ERANGE, 0, 0, 0);
+ errno = 0;
+ TEST_ff_f (pow, minus_zero, -0x1p127, plus_infty, DIVIDE_BY_ZERO_EXCEPTION);
+ check_int ("errno for pow(-0,-num) == ERANGE", errno, ERANGE, 0, 0, 0);
+ errno = 0;
+ /* Bug 13879: spurious OVERFLOW exception may be present. */
+ TEST_ff_f (pow, minus_zero, -max_value, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|OVERFLOW_EXCEPTION_OK);
check_int ("errno for pow(-0,-num) == ERANGE", errno, ERANGE, 0, 0, 0);
TEST_ff_f (pow, 0x1p72L, 0x1p72L, plus_infty, OVERFLOW_EXCEPTION);
@@ -5487,32 +5585,155 @@
/* pow (+inf, y) == +inf for y > 0. */
TEST_ff_f (pow, plus_infty, 2, plus_infty);
+ TEST_ff_f (pow, plus_infty, 0xffffff, plus_infty);
+#ifndef TEST_FLOAT
+ TEST_ff_f (pow, plus_infty, 0x1.fffffffffffffp+52L, plus_infty);
+#endif
+#ifdef TEST_LDOUBLE
+# if LDBL_MANT_DIG >= 64
+ TEST_ff_f (pow, plus_infty, 0x1.fffffffffffffffep+63L, plus_infty);
+# endif
+# if LDBL_MANT_DIG >= 106
+ TEST_ff_f (pow, plus_infty, 0x1.ffffffffffffffffffffffffff8p+105L, plus_infty);
+# endif
+# if LDBL_MANT_DIG >= 113
+ TEST_ff_f (pow, plus_infty, 0x1.ffffffffffffffffffffffffffffp+112L, plus_infty);
+# endif
+#endif
+ TEST_ff_f (pow, plus_infty, 0x1p24, plus_infty);
+ TEST_ff_f (pow, plus_infty, 0x1p127, plus_infty);
+ TEST_ff_f (pow, plus_infty, max_value, plus_infty);
/* pow (+inf, y) == +0 for y < 0. */
TEST_ff_f (pow, plus_infty, -1, 0.0);
+ TEST_ff_f (pow, plus_infty, -0xffffff, 0.0);
+#ifndef TEST_FLOAT
+ TEST_ff_f (pow, plus_infty, -0x1.fffffffffffffp+52L, 0.0);
+#endif
+#ifdef TEST_LDOUBLE
+# if LDBL_MANT_DIG >= 64
+ TEST_ff_f (pow, plus_infty, -0x1.fffffffffffffffep+63L, 0.0);
+# endif
+# if LDBL_MANT_DIG >= 106
+ TEST_ff_f (pow, plus_infty, -0x1.ffffffffffffffffffffffffff8p+105L, 0.0);
+# endif
+# if LDBL_MANT_DIG >= 113
+ TEST_ff_f (pow, plus_infty, -0x1.ffffffffffffffffffffffffffffp+112L, 0.0);
+# endif
+#endif
+ TEST_ff_f (pow, plus_infty, -0x1p24, 0.0);
+ TEST_ff_f (pow, plus_infty, -0x1p127, 0.0);
+ TEST_ff_f (pow, plus_infty, -max_value, 0.0);
/* pow (-inf, y) == -inf for y an odd integer > 0. */
TEST_ff_f (pow, minus_infty, 27, minus_infty);
+ TEST_ff_f (pow, minus_infty, 0xffffff, minus_infty);
+ TEST_ff_f (pow, minus_infty, 0x1fffffe, plus_infty);
+#ifndef TEST_FLOAT
+ TEST_ff_f (pow, minus_infty, 0x1.fffffffffffffp+52L, minus_infty);
+ TEST_ff_f (pow, minus_infty, 0x1.fffffffffffffp+53L, plus_infty);
+#endif
+#ifdef TEST_LDOUBLE
+# if LDBL_MANT_DIG >= 64
+ TEST_ff_f (pow, minus_infty, 0x1.fffffffffffffffep+63L, minus_infty);
+ TEST_ff_f (pow, minus_infty, 0x1.fffffffffffffffep+64L, plus_infty);
+# endif
+# if LDBL_MANT_DIG >= 106
+ TEST_ff_f (pow, minus_infty, 0x1.ffffffffffffffffffffffffff8p+105L, minus_infty);
+ TEST_ff_f (pow, minus_infty, 0x1.ffffffffffffffffffffffffff8p+106L, plus_infty);
+# endif
+# if LDBL_MANT_DIG >= 113
+ TEST_ff_f (pow, minus_infty, 0x1.ffffffffffffffffffffffffffffp+112L, minus_infty);
+ TEST_ff_f (pow, minus_infty, 0x1.ffffffffffffffffffffffffffffp+113L, plus_infty);
+# endif
+#endif
/* pow (-inf, y) == +inf for y > 0 and not an odd integer. */
TEST_ff_f (pow, minus_infty, 28, plus_infty);
+ TEST_ff_f (pow, minus_infty, 0x1p24, plus_infty);
+ TEST_ff_f (pow, minus_infty, 0x1p127, plus_infty);
+ TEST_ff_f (pow, minus_infty, max_value, plus_infty);
/* pow (-inf, y) == -0 for y an odd integer < 0. */
TEST_ff_f (pow, minus_infty, -3, minus_zero);
+ TEST_ff_f (pow, minus_infty, -0xffffff, minus_zero);
+ TEST_ff_f (pow, minus_infty, -0x1fffffe, plus_zero);
+#ifndef TEST_FLOAT
+ TEST_ff_f (pow, minus_infty, -0x1.fffffffffffffp+52L, minus_zero);
+ TEST_ff_f (pow, minus_infty, -0x1.fffffffffffffp+53L, plus_zero);
+#endif
+#ifdef TEST_LDOUBLE
+# if LDBL_MANT_DIG >= 64
+ TEST_ff_f (pow, minus_infty, -0x1.fffffffffffffffep+63L, minus_zero);
+ TEST_ff_f (pow, minus_infty, -0x1.fffffffffffffffep+64L, plus_zero);
+# endif
+# if LDBL_MANT_DIG >= 106
+ TEST_ff_f (pow, minus_infty, -0x1.ffffffffffffffffffffffffff8p+105L, minus_zero);
+ TEST_ff_f (pow, minus_infty, -0x1.ffffffffffffffffffffffffff8p+106L, plus_zero);
+# endif
+# if LDBL_MANT_DIG >= 113
+ TEST_ff_f (pow, minus_infty, -0x1.ffffffffffffffffffffffffffffp+112L, minus_zero);
+ TEST_ff_f (pow, minus_infty, -0x1.ffffffffffffffffffffffffffffp+113L, plus_zero);
+# endif
+#endif
/* pow (-inf, y) == +0 for y < 0 and not an odd integer. */
TEST_ff_f (pow, minus_infty, -2.0, 0.0);
+ TEST_ff_f (pow, minus_infty, -0x1p24, 0.0);
+ TEST_ff_f (pow, minus_infty, -0x1p127, 0.0);
+ TEST_ff_f (pow, minus_infty, -max_value, 0.0);
/* pow (+0, y) == +0 for y an odd integer > 0. */
TEST_ff_f (pow, 0.0, 27, 0.0);
+ TEST_ff_f (pow, 0.0, 0xffffff, 0.0);
+#ifndef TEST_FLOAT
+ TEST_ff_f (pow, 0.0, 0x1.fffffffffffffp+52L, 0.0);
+#endif
+#ifdef TEST_LDOUBLE
+# if LDBL_MANT_DIG >= 64
+ TEST_ff_f (pow, 0.0, 0x1.fffffffffffffffep+63L, 0.0);
+# endif
+# if LDBL_MANT_DIG >= 106
+ TEST_ff_f (pow, 0.0, 0x1.ffffffffffffffffffffffffff8p+105L, 0.0);
+# endif
+# if LDBL_MANT_DIG >= 113
+ TEST_ff_f (pow, 0.0, 0x1.ffffffffffffffffffffffffffffp+112L, 0.0);
+# endif
+#endif
/* pow (-0, y) == -0 for y an odd integer > 0. */
TEST_ff_f (pow, minus_zero, 27, minus_zero);
+ TEST_ff_f (pow, minus_zero, 0xffffff, minus_zero);
+ TEST_ff_f (pow, minus_zero, 0x1fffffe, plus_zero);
+#ifndef TEST_FLOAT
+ TEST_ff_f (pow, minus_zero, 0x1.fffffffffffffp+52L, minus_zero);
+ TEST_ff_f (pow, minus_zero, 0x1.fffffffffffffp+53L, plus_zero);
+#endif
+#ifdef TEST_LDOUBLE
+# if LDBL_MANT_DIG >= 64
+ TEST_ff_f (pow, minus_zero, 0x1.fffffffffffffffep+63L, minus_zero);
+ TEST_ff_f (pow, minus_zero, 0x1.fffffffffffffffep+64L, plus_zero);
+# endif
+# if LDBL_MANT_DIG >= 106
+ TEST_ff_f (pow, minus_zero, 0x1.ffffffffffffffffffffffffff8p+105L, minus_zero);
+ TEST_ff_f (pow, minus_zero, 0x1.ffffffffffffffffffffffffff8p+106L, plus_zero);
+# endif
+# if LDBL_MANT_DIG >= 113
+ TEST_ff_f (pow, minus_zero, 0x1.ffffffffffffffffffffffffffffp+112L, minus_zero);
+ TEST_ff_f (pow, minus_zero, 0x1.ffffffffffffffffffffffffffffp+113L, plus_zero);
+# endif
+#endif
/* pow (+0, y) == +0 for y > 0 and not an odd integer. */
TEST_ff_f (pow, 0.0, 4, 0.0);
+ TEST_ff_f (pow, 0.0, 0x1p24, 0.0);
+ TEST_ff_f (pow, 0.0, 0x1p127, 0.0);
+ TEST_ff_f (pow, 0.0, max_value, 0.0);
/* pow (-0, y) == +0 for y > 0 and not an odd integer. */
TEST_ff_f (pow, minus_zero, 4, 0.0);
+ TEST_ff_f (pow, minus_zero, 0x1p24, 0.0);
+ TEST_ff_f (pow, minus_zero, 0x1p127, 0.0);
+ TEST_ff_f (pow, minus_zero, max_value, 0.0);
TEST_ff_f (pow, 16, 0.25L, 2);
TEST_ff_f (pow, 0x1p64L, 0.125L, 256);
Modified: fsf/trunk/libc/math/w_exp2.c
==============================================================================
--- fsf/trunk/libc/math/w_exp2.c (original)
+++ fsf/trunk/libc/math/w_exp2.c Wed Mar 21 19:50:58 2012
@@ -2,23 +2,19 @@
* wrapper exp2(x)
*/
-#include <float.h>
#include <math.h>
#include <math_private.h>
-
-static const double o_threshold = (double) DBL_MAX_EXP;
-static const double u_threshold = (double) (DBL_MIN_EXP - DBL_MANT_DIG - 1);
double
__exp2 (double x)
{
- if (__builtin_expect (islessequal (x, u_threshold)
- || isgreater (x, o_threshold), 0)
- && _LIB_VERSION != _IEEE_ && __finite (x))
+ double z = __ieee754_exp2 (x);
+ if (__builtin_expect (!__finite (z), 0)
+ && __finite (x) && _LIB_VERSION != _IEEE_)
/* exp2 overflow: 44, exp2 underflow: 45 */
- return __kernel_standard (x, x, 44 + (x <= o_threshold));
+ return __kernel_standard (x, x, 44 + !!__signbit (x));
- return __ieee754_exp2 (x);
+ return z;
}
weak_alias (__exp2, exp2)
#ifdef NO_LONG_DOUBLE
Modified: fsf/trunk/libc/math/w_exp2f.c
==============================================================================
--- fsf/trunk/libc/math/w_exp2f.c (original)
+++ fsf/trunk/libc/math/w_exp2f.c Wed Mar 21 19:50:58 2012
@@ -2,22 +2,18 @@
* wrapper exp2f(x)
*/
-#include <float.h>
#include <math.h>
#include <math_private.h>
-
-static const float o_threshold = (float) FLT_MAX_EXP;
-static const float u_threshold = (float) (FLT_MIN_EXP - FLT_MANT_DIG - 1);
float
__exp2f (float x)
{
- if (__builtin_expect (islessequal (x, u_threshold)
- || isgreater (x, o_threshold), 0)
- && _LIB_VERSION != _IEEE_ && __finitef (x))
+ float z = __ieee754_exp2f (x);
+ if (__builtin_expect (!__finitef (z), 0)
+ && __finitef (x) && _LIB_VERSION != _IEEE_)
/* exp2 overflow: 144, exp2 underflow: 145 */
- return __kernel_standard_f (x, x, 144 + (x <= o_threshold));
+ return __kernel_standard_f (x, x, 144 + !!__signbitf (x));
- return __ieee754_exp2f (x);
+ return z;
}
weak_alias (__exp2f, exp2f)
Modified: fsf/trunk/libc/math/w_exp2l.c
==============================================================================
--- fsf/trunk/libc/math/w_exp2l.c (original)
+++ fsf/trunk/libc/math/w_exp2l.c Wed Mar 21 19:50:58 2012
@@ -2,23 +2,18 @@
* wrapper exp2l(x)
*/
-#include <float.h>
#include <math.h>
#include <math_private.h>
-
-static const long double o_threshold = (long double) LDBL_MAX_EXP;
-static const long double u_threshold
- = (long double) (LDBL_MIN_EXP - LDBL_MANT_DIG - 1);
long double
__exp2l (long double x)
{
- if (__builtin_expect (islessequal (x, u_threshold)
- || isgreater (x, o_threshold), 0)
- && _LIB_VERSION != _IEEE_ && __finitel (x))
+ long double z = __ieee754_exp2l (x);
+ if (__builtin_expect (!__finitel (z), 0)
+ && __finitel (x) && _LIB_VERSION != _IEEE_)
/* exp2 overflow: 244, exp2 underflow: 245 */
- return __kernel_standard (x, x, 244 + (x <= o_threshold));
+ return __kernel_standard (x, x, 244 + !!__signbitl (x));
- return __ieee754_exp2l (x);
+ return z;
}
weak_alias (__exp2l, exp2l)
Modified: fsf/trunk/libc/sysdeps/i386/configure
==============================================================================
--- fsf/trunk/libc/sysdeps/i386/configure (original)
+++ fsf/trunk/libc/sysdeps/i386/configure Wed Mar 21 19:50:58 2012
@@ -223,11 +223,12 @@
(eval $ac_try) 2>&5
ac_status=$?
$as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5
- test $ac_status = 0; }; }; then
+ test $ac_status = 0; }; }; then :
libc_cv_cc_sse4=yes
else
libc_cv_cc_sse4=no
fi
+
fi
{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $libc_cv_cc_sse4" >&5
$as_echo "$libc_cv_cc_sse4" >&6; }
@@ -246,11 +247,12 @@
(eval $ac_try) 2>&5
ac_status=$?
$as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5
- test $ac_status = 0; }; }; then
+ test $ac_status = 0; }; }; then :
libc_cv_as_i686=yes
else
libc_cv_as_i686=no
fi
+
fi
{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $libc_cv_as_i686" >&5
$as_echo "$libc_cv_as_i686" >&6; }
@@ -265,11 +267,12 @@
(eval $ac_try) 2>&5
ac_status=$?
$as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5
- test $ac_status = 0; }; }; then
+ test $ac_status = 0; }; }; then :
libc_cv_cc_avx=yes
else
libc_cv_cc_avx=no
fi
+
fi
{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $libc_cv_cc_avx" >&5
$as_echo "$libc_cv_cc_avx" >&6; }
@@ -288,11 +291,12 @@
(eval $ac_try) 2>&5
ac_status=$?
$as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5
- test $ac_status = 0; }; }; then
+ test $ac_status = 0; }; }; then :
libc_cv_cc_sse2avx=yes
else
libc_cv_cc_sse2avx=no
fi
+
fi
{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $libc_cv_cc_sse2avx" >&5
$as_echo "$libc_cv_cc_sse2avx" >&6; }
@@ -311,11 +315,12 @@
(eval $ac_try) 2>&5
ac_status=$?
$as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5
- test $ac_status = 0; }; }; then
+ test $ac_status = 0; }; }; then :
libc_cv_cc_fma4=yes
else
libc_cv_cc_fma4=no
fi
+
fi
{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $libc_cv_cc_fma4" >&5
$as_echo "$libc_cv_cc_fma4" >&6; }
@@ -334,11 +339,12 @@
(eval $ac_try) 2>&5
ac_status=$?
$as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5
- test $ac_status = 0; }; }; then
+ test $ac_status = 0; }; }; then :
libc_cv_cc_novzeroupper=yes
else
libc_cv_cc_novzeroupper=no
fi
+
fi
{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $libc_cv_cc_novzeroupper" >&5
$as_echo "$libc_cv_cc_novzeroupper" >&6; }
Modified: fsf/trunk/libc/sysdeps/i386/configure.in
==============================================================================
--- fsf/trunk/libc/sysdeps/i386/configure.in (original)
+++ fsf/trunk/libc/sysdeps/i386/configure.in Wed Mar 21 19:50:58 2012
@@ -40,60 +40,48 @@
dnl Check if -msse4 works.
AC_CACHE_CHECK(for SSE4 support, libc_cv_cc_sse4, [dnl
-if AC_TRY_COMMAND([${CC-cc} -msse4 -xc /dev/null -S -o /dev/null]); then
- libc_cv_cc_sse4=yes
-else
- libc_cv_cc_sse4=no
-fi])
+LIBC_TRY_CC_OPTION([-msse4], [libc_cv_cc_sse4=yes], [libc_cv_cc_sse4=no])
+])
if test $libc_cv_cc_sse4 = yes; then
AC_DEFINE(HAVE_SSE4_SUPPORT)
fi
dnl Check if -Wa,-mtune=i686 works.
AC_CACHE_CHECK(for assembler -mtune=i686 support, libc_cv_as_i686, [dnl
-if AC_TRY_COMMAND([${CC-cc} -Wa,-mtune=i686 -xc /dev/null -S -o /dev/null]); then
- libc_cv_as_i686=yes
-else
- libc_cv_as_i686=no
-fi])
+LIBC_TRY_CC_OPTION([-Wa,-mtune=i686],
+ [libc_cv_as_i686=yes],
+ [libc_cv_as_i686=no])
+])
dnl Check if -mavx works.
AC_CACHE_CHECK(for AVX support, libc_cv_cc_avx, [dnl
-if AC_TRY_COMMAND([${CC-cc} -mavx -xc /dev/null -S -o /dev/null]); then
- libc_cv_cc_avx=yes
-else
- libc_cv_cc_avx=no
-fi])
+LIBC_TRY_CC_OPTION([-mavx], [libc_cv_cc_avx=yes], [libc_cv_cc_avx=no])
+])
if test $libc_cv_cc_avx = yes; then
AC_DEFINE(HAVE_AVX_SUPPORT)
fi
dnl Check if -msse2avx works.
AC_CACHE_CHECK(for AVX encoding of SSE instructions, libc_cv_cc_sse2avx, [dnl
-if AC_TRY_COMMAND([${CC-cc} -msse2avx -xc /dev/null -S -o /dev/null]); then
- libc_cv_cc_sse2avx=yes
-else
- libc_cv_cc_sse2avx=no
-fi])
+LIBC_TRY_CC_OPTION([-msse2avx],
+ [libc_cv_cc_sse2avx=yes],
+ [libc_cv_cc_sse2avx=no])
+])
if test $libc_cv_cc_sse2avx = yes; then
AC_DEFINE(HAVE_SSE2AVX_SUPPORT)
fi
dnl Check if -mfma4 works.
AC_CACHE_CHECK(for FMA4 support, libc_cv_cc_fma4, [dnl
-if AC_TRY_COMMAND([${CC-cc} -mfma4 -xc /dev/null -S -o /dev/null]); then
- libc_cv_cc_fma4=yes
-else
- libc_cv_cc_fma4=no
-fi])
+LIBC_TRY_CC_OPTION([-mfma4], [libc_cv_cc_fma4=yes], [libc_cv_cc_fma4=no])
+])
if test $libc_cv_cc_fma4 = yes; then
AC_DEFINE(HAVE_FMA4_SUPPORT)
fi
dnl Check if -mno-vzeroupper works.
AC_CACHE_CHECK(for -mno-vzeroupper support, libc_cv_cc_novzeroupper, [dnl
-if AC_TRY_COMMAND([${CC-cc} -mno-vzeroupper -xc /dev/null -S -o /dev/null]); then
- libc_cv_cc_novzeroupper=yes
-else
- libc_cv_cc_novzeroupper=no
-fi])
+LIBC_TRY_CC_OPTION([-mno-vzeroupper],
+ [libc_cv_cc_novzeroupper=yes],
+ [libc_cv_cc_novzeroupper=no])
+])
Modified: fsf/trunk/libc/sysdeps/i386/fpu/e_pow.S
==============================================================================
--- fsf/trunk/libc/sysdeps/i386/fpu/e_pow.S (original)
+++ fsf/trunk/libc/sysdeps/i386/fpu/e_pow.S Wed Mar 21 19:50:58 2012
@@ -230,6 +230,16 @@
testb $2, %dh
jz 16f // jump if x == +inf
+ // fistpll raises invalid exception for |y| >= 1L<<63, so test
+ // that (in which case y is certainly even) before testing
+ // whether y is odd.
+ fld %st // y : y
+ fabs // |y| : y
+ fcompl MO(p63) // y
+ fnstsw
+ sahf
+ jnc 16f
+
// We must find out whether y is an odd integer.
fld %st // y : y
fistpll (%esp) // y
@@ -239,20 +249,13 @@
sahf
jne 17f
- // OK, the value is an integer, but is the number of bits small
- // enough so that all are coming from the mantissa?
+ // OK, the value is an integer.
popl %eax
cfi_adjust_cfa_offset (-4)
popl %edx
cfi_adjust_cfa_offset (-4)
andb $1, %al
jz 18f // jump if not odd
- movl %edx, %eax
- orl %edx, %edx
- jns 155f
- negl %eax
-155: cmpl $0x00200000, %eax
- ja 18f // does not fit in mantissa bits
// It's an odd integer.
shrl $31, %edx
fldl MOX(minf_mzero, %edx, 8)
@@ -289,6 +292,16 @@
testb $2, %dh
jz 25f
+ // fistpll raises invalid exception for |y| >= 1L<<63, so test
+ // that (in which case y is certainly even) before testing
+ // whether y is odd.
+ fld %st // y : y
+ fabs // |y| : y
+ fcompl MO(p63) // y
+ fnstsw
+ sahf
+ jnc 25f
+
fld %st // y : y
fistpll (%esp) // y
fildll (%esp) // int(y) : y
@@ -297,16 +310,13 @@
sahf
jne 26f
- // OK, the value is an integer, but is the number of bits small
- // enough so that all are coming from the mantissa?
+ // OK, the value is an integer.
popl %eax
cfi_adjust_cfa_offset (-4)
popl %edx
cfi_adjust_cfa_offset (-4)
andb $1, %al
jz 27f // jump if not odd
- cmpl $0xffe00000, %edx
- jbe 27f // does not fit in mantissa bits
// It's an odd integer.
// Raise divide-by-zero exception and get minus infinity value.
fldl MO(one)
@@ -329,6 +339,14 @@
21: testb $2, %dh
jz 22f
+ // fistpll raises invalid exception for |y| >= 1L<<63, so test
+ // that (in which case y is certainly even) before testing
+ // whether y is odd.
+ fcoml MO(p63) // y
+ fnstsw
+ sahf
+ jnc 22f
+
fld %st // y : y
fistpll (%esp) // y
fildll (%esp) // int(y) : y
@@ -337,16 +355,13 @@
sahf
jne 23f
- // OK, the value is an integer, but is the number of bits small
- // enough so that all are coming from the mantissa?
+ // OK, the value is an integer.
popl %eax
cfi_adjust_cfa_offset (-4)
popl %edx
cfi_adjust_cfa_offset (-4)
andb $1, %al
jz 24f // jump if not odd
- cmpl $0xffe00000, %edx
- jae 24f // does not fit in mantissa bits
// It's an odd integer.
fldl MO(mzero)
ret
Modified: fsf/trunk/libc/sysdeps/i386/fpu/e_powf.S
==============================================================================
--- fsf/trunk/libc/sysdeps/i386/fpu/e_powf.S (original)
+++ fsf/trunk/libc/sysdeps/i386/fpu/e_powf.S Wed Mar 21 19:50:58 2012
@@ -224,6 +224,16 @@
testb $2, %dh
jz 16f // jump if x == +inf
+ // fistpl raises invalid exception for |y| >= 1L<<31, so test
+ // that (in which case y is certainly even) before testing
+ // whether y is odd.
+ fld %st // y : y
+ fabs // |y| : y
+ fcompl MO(p31) // y
+ fnstsw
+ sahf
+ jnc 16f
+
// We must find out whether y is an odd integer.
fld %st // y : y
fistpl (%esp) // y
@@ -233,18 +243,11 @@
sahf
jne 17f
- // OK, the value is an integer, but is the number of bits small
- // enough so that all are coming from the mantissa?
+ // OK, the value is an integer.
popl %edx
cfi_adjust_cfa_offset (-4)
testb $1, %dl
jz 18f // jump if not odd
- movl %edx, %eax
- orl %edx, %edx
- jns 155f
- negl %eax
-155: cmpl $0x01000000, %eax
- ja 18f // does not fit in mantissa bits
// It's an odd integer.
shrl $31, %edx
fldl MOX(minf_mzero, %edx, 8)
@@ -281,6 +284,16 @@
testb $2, %dh
jz 25f
+ // fistpl raises invalid exception for |y| >= 1L<<31, so test
+ // that (in which case y is certainly even) before testing
+ // whether y is odd.
+ fld %st // y : y
+ fabs // |y| : y
+ fcompl MO(p31) // y
+ fnstsw
+ sahf
+ jnc 25f
+
fld %st // y : y
fistpl (%esp) // y
fildl (%esp) // int(y) : y
@@ -289,14 +302,11 @@
sahf
jne 26f
- // OK, the value is an integer, but is the number of bits small
- // enough so that all are coming from the mantissa?
+ // OK, the value is an integer.
popl %edx
cfi_adjust_cfa_offset (-4)
testb $1, %dl
jz 27f // jump if not odd
- cmpl $0xff000000, %edx
- jbe 27f // does not fit in mantissa bits
// It's an odd integer.
// Raise divide-by-zero exception and get minus infinity value.
fldl MO(one)
@@ -319,6 +329,14 @@
21: testb $2, %dh
jz 22f
+ // fistpl raises invalid exception for |y| >= 1L<<31, so test
+ // that (in which case y is certainly even) before testing
+ // whether y is odd.
+ fcoml MO(p31) // y
+ fnstsw
+ sahf
+ jnc 22f
+
fld %st // y : y
fistpl (%esp) // y
fildl (%esp) // int(y) : y
@@ -327,14 +345,11 @@
sahf
jne 23f
- // OK, the value is an integer, but is the number of bits small
- // enough so that all are coming from the mantissa?
+ // OK, the value is an integer.
popl %edx
cfi_adjust_cfa_offset (-4)
testb $1, %dl
jz 24f // jump if not odd
- cmpl $0xff000000, %edx
- jae 24f // does not fit in mantissa bits
// It's an odd integer.
fldl MO(mzero)
ret
Modified: fsf/trunk/libc/sysdeps/i386/fpu/e_powl.S
==============================================================================
--- fsf/trunk/libc/sysdeps/i386/fpu/e_powl.S (original)
+++ fsf/trunk/libc/sysdeps/i386/fpu/e_powl.S Wed Mar 21 19:50:58 2012
@@ -32,6 +32,9 @@
ASM_TYPE_DIRECTIVE(p63,@object)
p63: .byte 0, 0, 0, 0, 0, 0, 0xe0, 0x43
ASM_SIZE_DIRECTIVE(p63)
+ ASM_TYPE_DIRECTIVE(p64,@object)
+p64: .byte 0, 0, 0, 0, 0, 0, 0xf0, 0x43
+ ASM_SIZE_DIRECTIVE(p64)
.section .rodata.cst16,"aM",@progbits,16
@@ -243,6 +246,19 @@
testb $2, %dh
jz 16f // jump if x == +inf
+ // fistpll raises invalid exception for |y| >= 1L<<63, but y
+ // may be odd unless we know |y| >= 1L<<64.
+ fld %st // y : y
+ fabs // |y| : y
+ fcompl MO(p64) // y
+ fnstsw
+ sahf
+ jnc 16f
+ fldl MO(p63) // p63 : y
+ fxch // y : p63
+ fprem // y%p63 : p63
+ fstp %st(1) // y%p63
+
// We must find out whether y is an odd integer.
fld %st // y : y
fistpll (%esp) // y
@@ -294,6 +310,19 @@
// x is Â0 and y is < 0. We must find out whether y is an odd integer.
testb $2, %dh
jz 25f
+
+ // fistpll raises invalid exception for |y| >= 1L<<63, but y
+ // may be odd unless we know |y| >= 1L<<64.
+ fld %st // y : y
+ fabs // |y| : y
+ fcompl MO(p64) // y
+ fnstsw
+ sahf
+ jnc 25f
+ fldl MO(p63) // p63 : y
+ fxch // y : p63
+ fprem // y%p63 : p63
+ fstp %st(1) // y%p63
fld %st // y : y
fistpll (%esp) // y
@@ -332,6 +361,18 @@
21: testb $2, %dh
jz 22f
+ // fistpll raises invalid exception for |y| >= 1L<<63, but y
+ // may be odd unless we know |y| >= 1L<<64.
+ fld %st // y : y
+ fcompl MO(p64) // y
+ fnstsw
+ sahf
+ jnc 22f
+ fldl MO(p63) // p63 : y
+ fxch // y : p63
+ fprem // y%p63 : p63
+ fstp %st(1) // y%p63
+
fld %st // y : y
fistpll (%esp) // y
fildll (%esp) // int(y) : y
Modified: fsf/trunk/libc/sysdeps/i386/fpu/libm-test-ulps
==============================================================================
--- fsf/trunk/libc/sysdeps/i386/fpu/libm-test-ulps (original)
+++ fsf/trunk/libc/sysdeps/i386/fpu/libm-test-ulps Wed Mar 21 19:50:58 2012
@@ -431,15 +431,44 @@
ifloat: 1
# cexp
+Test "Real part of: cexp (-10000 + 0x1p16383 i) == 1.045876464564882298442774542991176546722e-4343 + 4.421154026488516836023811173959413420548e-4344 i":
+ildouble: 1
+ldouble: 1
Test "Real part of: cexp (-2.0 - 3.0 i) == -0.13398091492954261346140525546115575 - 0.019098516261135196432576240858800925 i":
ildouble: 1
ldouble: 1
Test "Imaginary part of: cexp (-2.0 - 3.0 i) == -0.13398091492954261346140525546115575 - 0.019098516261135196432576240858800925 i":
-ildouble: 1
-ldouble: 1
+float: 1
+ifloat: 1
+ildouble: 1
+ldouble: 1
+Test "Imaginary part of: cexp (0 + 0x1p65 i) == 0.99888622066058013610642172179340364209972 - 0.047183876212354673805106149805700013943218 i":
+float: 1
+ifloat: 1
+Test "Imaginary part of: cexp (0 - 0x1p65 i) == 0.99888622066058013610642172179340364209972 + 0.047183876212354673805106149805700013943218 i":
+float: 1
+ifloat: 1
Test "Real part of: cexp (0.75 + 1.25 i) == 0.667537446429131586942201977015932112 + 2.00900045494094876258347228145863909 i":
-ildouble: 1
-ldouble: 1
+float: 1
+ifloat: 1
+ildouble: 1
+ldouble: 1
+Test "Imaginary part of: cexp (0.75 + 1.25 i) == 0.667537446429131586942201977015932112 + 2.00900045494094876258347228145863909 i":
+ildouble: 1
+ldouble: 1
+Test "Real part of: cexp (50 + 0x1p127 i) == 4.053997150228616856622417636046265337193e21 + 3.232070315463388524466674772633810238819e21 i":
+double: 2
+idouble: 2
+Test "Imaginary part of: cexp (50 + 0x1p127 i) == 4.053997150228616856622417636046265337193e21 + 3.232070315463388524466674772633810238819e21 i":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+ildouble: 1
+ldouble: 1
+Test "Real part of: cexp (500 + 0x1p1023 i) == -1.159886268932754433233243794561351783426e217 + 7.904017694554466595359379965081774849708e216 i":
+double: 1
+idouble: 1
# clog
Test "Real part of: clog (0.75 + 1.25 i) == 0.376885901188190075998919126749298416 + 1.03037682652431246378774332703115153 i":
@@ -815,18 +844,22 @@
ildouble: 1
ldouble: 1
Test "Imaginary part of: cpow (0.75 + 1.25 i, 0.0 + 1.0 i) == 0.331825439177608832276067945276730566 + 0.131338600281188544930936345230903032 i":
+float: 1
+ifloat: 1
ildouble: 1
ldouble: 1
Test "Real part of: cpow (0.75 + 1.25 i, 0.75 + 1.25 i) == 0.117506293914473555420279832210420483 + 0.346552747708338676483025352060418001 i":
-float: 3
-ifloat: 3
+double: 1
+float: 4
+idouble: 1
+ifloat: 4
ildouble: 6
ldouble: 6
Test "Imaginary part of: cpow (0.75 + 1.25 i, 0.75 + 1.25 i) == 0.117506293914473555420279832210420483 + 0.346552747708338676483025352060418001 i":
float: 1
ifloat: 1
-ildouble: 1
-ldouble: 1
+ildouble: 2
+ldouble: 2
Test "Real part of: cpow (0.75 + 1.25 i, 1.0 + 0.0 i) == 0.75 + 1.25 i":
ildouble: 1
ldouble: 1
@@ -834,22 +867,27 @@
float: 1
ifloat: 1
Test "Real part of: cpow (0.75 + 1.25 i, 1.0 + 1.0 i) == 0.0846958290317209430433805274189191353 + 0.513285749182902449043287190519090481 i":
-double: 1
+double: 2
float: 3
-idouble: 1
+idouble: 2
ifloat: 3
ildouble: 3
ldouble: 3
+Test "Real part of: cpow (2 + 0 i, 10 + 0 i) == 1024.0 + 0.0 i":
+ildouble: 1
+ldouble: 1
Test "Real part of: cpow (2 + 3 i, 4 + 0 i) == -119.0 - 120.0 i":
double: 1
-float: 4
-idouble: 1
-ifloat: 4
+float: 5
+idouble: 1
+ifloat: 5
+ildouble: 1
+ldouble: 1
Test "Imaginary part of: cpow (2 + 3 i, 4 + 0 i) == -119.0 - 120.0 i":
-float: 1
-ifloat: 1
-ildouble: 2
-ldouble: 2
+float: 2
+ifloat: 2
+ildouble: 4
+ldouble: 4
Test "Imaginary part of: cpow (e + 0 i, 0 + 2 * M_PIl i) == 1.0 + 0.0 i":
double: 2
float: 3
@@ -2124,10 +2162,18 @@
ldouble: 1
Function: Real part of "cexp":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
ildouble: 1
ldouble: 1
Function: Imaginary part of "cexp":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
ildouble: 1
ldouble: 1
@@ -2212,20 +2258,20 @@
ldouble: 1
Function: Real part of "cpow":
-double: 1
-float: 4
-idouble: 1
-ifloat: 4
-ildouble: 6
-ldouble: 6
+double: 2
+float: 5
+idouble: 2
+ifloat: 5
+ildouble: 5
+ldouble: 5
Function: Imaginary part of "cpow":
double: 2
float: 3
idouble: 2
ifloat: 3
-ildouble: 2
-ldouble: 2
+ildouble: 4
+ldouble: 4
Function: Real part of "csin":
float: 1
Removed: fsf/trunk/libc/sysdeps/i386/fpu/s_cexp.S
==============================================================================
--- fsf/trunk/libc/sysdeps/i386/fpu/s_cexp.S (original)
+++ fsf/trunk/libc/sysdeps/i386/fpu/s_cexp.S (removed)
@@ -1,253 +1,0 @@
-/* ix87 specific implementation of complex exponential function for double.
- Copyright (C) 1997, 2005, 2012 Free Software Foundation, Inc.
- This file is part of the GNU C Library.
- Contributed by Ulrich Drepper <drepper@xxxxxxxxxx>, 1997.
-
- 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 <sysdep.h>
-
- .section .rodata
-
- .align ALIGNARG(4)
- ASM_TYPE_DIRECTIVE(huge_nan_null_null,@object)
-huge_nan_null_null:
- .byte 0, 0, 0, 0, 0, 0, 0xf0, 0x7f
- .byte 0, 0, 0, 0, 0, 0, 0xff, 0x7f
- .double 0.0
-zero: .double 0.0
-infinity:
- .byte 0, 0, 0, 0, 0, 0, 0xf0, 0x7f
- .byte 0, 0, 0, 0, 0, 0, 0xff, 0x7f
- .double 0.0
- .byte 0, 0, 0, 0, 0, 0, 0, 0x80
- ASM_SIZE_DIRECTIVE(huge_nan_null_null)
-
- ASM_TYPE_DIRECTIVE(twopi,@object)
-twopi:
- .byte 0x35, 0xc2, 0x68, 0x21, 0xa2, 0xda, 0xf, 0xc9, 0x1, 0x40
- .byte 0, 0, 0, 0, 0, 0
- ASM_SIZE_DIRECTIVE(twopi)
-
- ASM_TYPE_DIRECTIVE(l2e,@object)
-l2e:
- .byte 0xbc, 0xf0, 0x17, 0x5c, 0x29, 0x3b, 0xaa, 0xb8, 0xff, 0x3f
- .byte 0, 0, 0, 0, 0, 0
- ASM_SIZE_DIRECTIVE(l2e)
-
- ASM_TYPE_DIRECTIVE(one,@object)
-one: .double 1.0
- ASM_SIZE_DIRECTIVE(one)
-
-
-#ifdef PIC
-#define MO(op) op##@GOTOFF(%ecx)
-#define MOX(op,x,f) op##@GOTOFF(%ecx,x,f)
-#else
-#define MO(op) op
-#define MOX(op,x,f) op(,x,f)
-#endif
-
- .text
-ENTRY(__cexp)
- fldl 8(%esp) /* x */
- fxam
- fnstsw
- fldl 16(%esp) /* y : x */
-#ifdef PIC
- LOAD_PIC_REG (cx)
-#endif
- movb %ah, %dh
- andb $0x45, %ah
- cmpb $0x05, %ah
- je 1f /* Jump if real part is +-Inf */
- cmpb $0x01, %ah
- je 2f /* Jump if real part is NaN */
-
- fxam /* y : x */
- fnstsw
- /* If the imaginary part is not finite we return NaN+i NaN, as
- for the case when the real part is NaN. A test for +-Inf and
- NaN would be necessary. But since we know the stack register
- we applied `fxam' to is not empty we can simply use one test.
- Check your FPU manual for more information. */
- andb $0x01, %ah
- cmpb $0x01, %ah
- je 20f
-
- /* We have finite numbers in the real and imaginary part. Do
- the real work now. */
- fxch /* x : y */
- fldt MO(l2e) /* log2(e) : x : y */
- fmulp /* x * log2(e) : y */
- fld %st /* x * log2(e) : x * log2(e) : y */
- frndint /* int(x * log2(e)) : x * log2(e) : y */
- fsubr %st, %st(1) /* int(x * log2(e)) : frac(x * log2(e)) : y */
- fxch /* frac(x * log2(e)) : int(x * log2(e)) : y */
- f2xm1 /* 2^frac(x * log2(e))-1 : int(x * log2(e)) : y */
- faddl MO(one) /* 2^frac(x * log2(e)) : int(x * log2(e)) : y */
- fscale /* e^x : int(x * log2(e)) : y */
- fst %st(1) /* e^x : e^x : y */
- fxch %st(2) /* y : e^x : e^x */
- fsincos /* cos(y) : sin(y) : e^x : e^x */
- fnstsw
- testl $0x400, %eax
- jnz 7f
- fmulp %st, %st(3) /* sin(y) : e^x : e^x * cos(y) */
- fmulp %st, %st(1) /* e^x * sin(y) : e^x * cos(y) */
- movl 4(%esp), %eax /* Pointer to memory for result. */
- fstpl 8(%eax)
- fstpl (%eax)
- ret $4
-
- /* We have to reduce the argument to fsincos. */
- .align ALIGNARG(4)
-7: fldt MO(twopi) /* 2*pi : y : e^x : e^x */
- fxch /* y : 2*pi : e^x : e^x */
-8: fprem1 /* y%(2*pi) : 2*pi : e^x : e^x */
- fnstsw
- testl $0x400, %eax
- jnz 8b
- fstp %st(1) /* y%(2*pi) : e^x : e^x */
- fsincos /* cos(y) : sin(y) : e^x : e^x */
- fmulp %st, %st(3)
- fmulp %st, %st(1)
- movl 4(%esp), %eax /* Pointer to memory for result. */
- fstpl 8(%eax)
- fstpl (%eax)
- ret $4
-
- /* The real part is +-inf. We must make further differences. */
- .align ALIGNARG(4)
-1: fxam /* y : x */
- fnstsw
- movb %ah, %dl
- testb $0x01, %ah /* See above why 0x01 is usable here. */
- jne 3f
-
-
- /* The real part is +-Inf and the imaginary part is finite. */
- andl $0x245, %edx
- cmpb $0x40, %dl /* Imaginary part == 0? */
- je 4f /* Yes -> */
-
- fxch /* x : y */
- shrl $5, %edx
- fstp %st(0) /* y */ /* Drop the real part. */
- andl $16, %edx /* This puts the sign bit of the real part
- in bit 4. So we can use it to index a
- small array to select 0 or Inf. */
- fsincos /* cos(y) : sin(y) */
- fnstsw
- testl $0x0400, %eax
- jnz 5f
- fldl MOX(huge_nan_null_null,%edx,1)
- movl 4(%esp), %edx /* Pointer to memory for result. */
- fstl 8(%edx)
- fstpl (%edx)
- ftst
- fnstsw
- shll $23, %eax
- andl $0x80000000, %eax
- orl %eax, 4(%edx)
- fstp %st(0)
- ftst
- fnstsw
- shll $23, %eax
- andl $0x80000000, %eax
- orl %eax, 12(%edx)
- fstp %st(0)
- ret $4
- /* We must reduce the argument to fsincos. */
- .align ALIGNARG(4)
-5: fldt MO(twopi)
- fxch
-6: fprem1
- fnstsw
- testl $0x400, %eax
- jnz 6b
- fstp %st(1)
- fsincos
- fldl MOX(huge_nan_null_null,%edx,1)
- movl 4(%esp), %edx /* Pointer to memory for result. */
- fstl 8(%edx)
- fstpl (%edx)
- ftst
- fnstsw
- shll $23, %eax
- andl $0x80000000, %eax
- orl %eax, 4(%edx)
- fstp %st(0)
- ftst
- fnstsw
- shll $23, %eax
- andl $0x80000000, %eax
- orl %eax, 12(%edx)
- fstp %st(0)
- ret $4
-
- /* The real part is +-Inf and the imaginary part is +-0. So return
- +-Inf+-0i. */
- .align ALIGNARG(4)
-4: movl 4(%esp), %eax /* Pointer to memory for result. */
- fstpl 8(%eax)
- shrl $5, %edx
- fstp %st(0)
- andl $16, %edx
- fldl MOX(huge_nan_null_null,%edx,1)
- fstpl (%eax)
- ret $4
-
- /* The real part is +-Inf, the imaginary is also is not finite. */
- .align ALIGNARG(4)
-3: fstp %st(0)
- fstp %st(0) /* <empty> */
- andb $0x45, %ah
- andb $0x47, %dh
- xorb %dh, %ah
- jnz 30f
- fldl MO(infinity) /* Raise invalid exception. */
- fmull MO(zero)
- fstp %st(0)
-30: movl %edx, %eax
- shrl $5, %edx
- shll $4, %eax
- andl $16, %edx
- andl $32, %eax
- orl %eax, %edx
- movl 4(%esp), %eax /* Pointer to memory for result. */
-
- fldl MOX(huge_nan_null_null,%edx,1)
- fldl MOX(huge_nan_null_null+8,%edx,1)
- fxch
- fstpl (%eax)
- fstpl 8(%eax)
- ret $4
-
- /* The real part is NaN. */
- .align ALIGNARG(4)
-20: fldl MO(infinity) /* Raise invalid exception. */
- fmull MO(zero)
- fstp %st(0)
-2: fstp %st(0)
- fstp %st(0)
- movl 4(%esp), %eax /* Pointer to memory for result. */
- fldl MO(huge_nan_null_null+8)
- fstl (%eax)
- fstpl 8(%eax)
- ret $4
-
-END(__cexp)
-weak_alias (__cexp, cexp)
Removed: fsf/trunk/libc/sysdeps/i386/fpu/s_cexpf.S
==============================================================================
--- fsf/trunk/libc/sysdeps/i386/fpu/s_cexpf.S (original)
+++ fsf/trunk/libc/sysdeps/i386/fpu/s_cexpf.S (removed)
@@ -1,257 +1,0 @@
-/* ix87 specific implementation of complex exponential function for double.
- Copyright (C) 1997, 2005, 2012 Free Software Foundation, Inc.
- This file is part of the GNU C Library.
- Contributed by Ulrich Drepper <drepper@xxxxxxxxxx>, 1997.
-
- 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 <sysdep.h>
-
- .section .rodata
-
- .align ALIGNARG(4)
- ASM_TYPE_DIRECTIVE(huge_nan_null_null,@object)
-huge_nan_null_null:
- .byte 0, 0, 0x80, 0x7f
- .byte 0, 0, 0xc0, 0x7f
- .float 0.0
-zero: .float 0.0
-infinity:
- .byte 0, 0, 0x80, 0x7f
- .byte 0, 0, 0xc0, 0x7f
- .float 0.0
- .byte 0, 0, 0, 0x80
- ASM_SIZE_DIRECTIVE(huge_nan_null_null)
-
- ASM_TYPE_DIRECTIVE(twopi,@object)
-twopi:
- .byte 0x35, 0xc2, 0x68, 0x21, 0xa2, 0xda, 0xf, 0xc9, 0x1, 0x40
- .byte 0, 0, 0, 0, 0, 0
- ASM_SIZE_DIRECTIVE(twopi)
-
- ASM_TYPE_DIRECTIVE(l2e,@object)
-l2e:
- .byte 0xbc, 0xf0, 0x17, 0x5c, 0x29, 0x3b, 0xaa, 0xb8, 0xff, 0x3f
- .byte 0, 0, 0, 0, 0, 0
- ASM_SIZE_DIRECTIVE(l2e)
-
- ASM_TYPE_DIRECTIVE(one,@object)
-one: .double 1.0
- ASM_SIZE_DIRECTIVE(one)
-
-
-#ifdef PIC
-#define MO(op) op##@GOTOFF(%ecx)
-#define MOX(op,x,f) op##@GOTOFF(%ecx,x,f)
-#else
-#define MO(op) op
-#define MOX(op,x,f) op(,x,f)
-#endif
-
- .text
-ENTRY(__cexpf)
- flds 4(%esp) /* x */
- fxam
- fnstsw
- flds 8(%esp) /* y : x */
-#ifdef PIC
- LOAD_PIC_REG (cx)
-#endif
- movb %ah, %dh
- andb $0x45, %ah
- cmpb $0x05, %ah
- je 1f /* Jump if real part is +-Inf */
- cmpb $0x01, %ah
- je 2f /* Jump if real part is NaN */
-
- fxam /* y : x */
- fnstsw
- /* If the imaginary part is not finite we return NaN+i NaN, as
- for the case when the real part is NaN. A test for +-Inf and
- NaN would be necessary. But since we know the stack register
- we applied `fxam' to is not empty we can simply use one test.
- Check your FPU manual for more information. */
- andb $0x01, %ah
- cmpb $0x01, %ah
- je 20f
-
- /* We have finite numbers in the real and imaginary part. Do
- the real work now. */
- fxch /* x : y */
- fldt MO(l2e) /* log2(e) : x : y */
- fmulp /* x * log2(e) : y */
- fld %st /* x * log2(e) : x * log2(e) : y */
- frndint /* int(x * log2(e)) : x * log2(e) : y */
- fsubr %st, %st(1) /* int(x * log2(e)) : frac(x * log2(e)) : y */
- fxch /* frac(x * log2(e)) : int(x * log2(e)) : y */
- f2xm1 /* 2^frac(x * log2(e))-1 : int(x * log2(e)) : y */
- faddl MO(one) /* 2^frac(x * log2(e)) : int(x * log2(e)) : y */
- fscale /* e^x : int(x * log2(e)) : y */
- fst %st(1) /* e^x : e^x : y */
- fxch %st(2) /* y : e^x : e^x */
- fsincos /* cos(y) : sin(y) : e^x : e^x */
- fnstsw
- testl $0x400, %eax
- jnz 7f
- fmulp %st, %st(3) /* sin(y) : e^x : e^x * cos(y) */
- fmulp %st, %st(1) /* e^x * sin(y) : e^x * cos(y) */
- subl $8, %esp
- cfi_adjust_cfa_offset (8)
- fstps 4(%esp)
- fstps (%esp)
- popl %eax
- cfi_adjust_cfa_offset (-4)
- popl %edx
- cfi_adjust_cfa_offset (-4)
- ret
-
- /* We have to reduce the argument to fsincos. */
- .align ALIGNARG(4)
-7: fldt MO(twopi) /* 2*pi : y : e^x : e^x */
- fxch /* y : 2*pi : e^x : e^x */
-8: fprem1 /* y%(2*pi) : 2*pi : e^x : e^x */
- fnstsw
- testl $0x400, %eax
- jnz 8b
- fstp %st(1) /* y%(2*pi) : e^x : e^x */
- fsincos /* cos(y) : sin(y) : e^x : e^x */
- fmulp %st, %st(3)
- fmulp %st, %st(1)
- subl $8, %esp
- cfi_adjust_cfa_offset (8)
- fstps 4(%esp)
- fstps (%esp)
- popl %eax
- cfi_adjust_cfa_offset (-4)
- popl %edx
- cfi_adjust_cfa_offset (-4)
- ret
-
- /* The real part is +-inf. We must make further differences. */
- .align ALIGNARG(4)
-1: fxam /* y : x */
- fnstsw
- movb %ah, %dl
- testb $0x01, %ah /* See above why 0x01 is usable here. */
- jne 3f
-
-
- /* The real part is +-Inf and the imaginary part is finite. */
- andl $0x245, %edx
- cmpb $0x40, %dl /* Imaginary part == 0? */
- je 4f /* Yes -> */
-
- fxch /* x : y */
- shrl $6, %edx
- fstp %st(0) /* y */ /* Drop the real part. */
- andl $8, %edx /* This puts the sign bit of the real part
- in bit 3. So we can use it to index a
- small array to select 0 or Inf. */
- fsincos /* cos(y) : sin(y) */
- fnstsw
- testl $0x0400, %eax
- jnz 5f
- fxch
- ftst
- fnstsw
- fstp %st(0)
- shll $23, %eax
- andl $0x80000000, %eax
- orl MOX(huge_nan_null_null,%edx,1), %eax
- movl MOX(huge_nan_null_null,%edx,1), %ecx
- movl %eax, %edx
- ftst
- fnstsw
- fstp %st(0)
- shll $23, %eax
- andl $0x80000000, %eax
- orl %ecx, %eax
- ret
- /* We must reduce the argument to fsincos. */
- .align ALIGNARG(4)
-5: fldt MO(twopi)
- fxch
-6: fprem1
- fnstsw
- testl $0x400, %eax
- jnz 6b
- fstp %st(1)
- fsincos
- fxch
- ftst
- fnstsw
- fstp %st(0)
- shll $23, %eax
- andl $0x80000000, %eax
- orl MOX(huge_nan_null_null,%edx,1), %eax
- movl MOX(huge_nan_null_null,%edx,1), %ecx
- movl %eax, %edx
- ftst
- fnstsw
- fstp %st(0)
- shll $23, %eax
- andl $0x80000000, %eax
- orl %ecx, %eax
- ret
-
- /* The real part is +-Inf and the imaginary part is +-0. So return
- +-Inf+-0i. */
- .align ALIGNARG(4)
-4: subl $4, %esp
- cfi_adjust_cfa_offset (4)
- fstps (%esp)
- shrl $6, %edx
- fstp %st(0)
- andl $8, %edx
- movl MOX(huge_nan_null_null,%edx,1), %eax
- popl %edx
- cfi_adjust_cfa_offset (-4)
- ret
-
- /* The real part is +-Inf, the imaginary is also is not finite. */
- .align ALIGNARG(4)
-3: fstp %st(0)
- fstp %st(0) /* <empty> */
- andb $0x45, %ah
- andb $0x47, %dh
- xorb %dh, %ah
- jnz 30f
- flds MO(infinity) /* Raise invalid exception. */
- fmuls MO(zero)
- fstp %st(0)
-30: movl %edx, %eax
- shrl $6, %edx
- shll $3, %eax
- andl $8, %edx
- andl $16, %eax
- orl %eax, %edx
-
- movl MOX(huge_nan_null_null,%edx,1), %eax
- movl MOX(huge_nan_null_null+4,%edx,1), %edx
- ret
-
- /* The real part is NaN. */
- .align ALIGNARG(4)
-20: flds MO(infinity) /* Raise invalid exception. */
- fmuls MO(zero)
- fstp %st(0)
-2: fstp %st(0)
- fstp %st(0)
- movl MO(huge_nan_null_null+4), %eax
- movl %eax, %edx
- ret
-
-END(__cexpf)
-weak_alias (__cexpf, cexpf)
Removed: fsf/trunk/libc/sysdeps/i386/fpu/s_cexpl.S
==============================================================================
--- fsf/trunk/libc/sysdeps/i386/fpu/s_cexpl.S (original)
+++ fsf/trunk/libc/sysdeps/i386/fpu/s_cexpl.S (removed)
@@ -1,256 +1,0 @@
-/* ix87 specific implementation of complex exponential function for double.
- Copyright (C) 1997, 2005, 2012 Free Software Foundation, Inc.
- This file is part of the GNU C Library.
- Contributed by Ulrich Drepper <drepper@xxxxxxxxxx>, 1997.
-
- 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 <sysdep.h>
-
- .section .rodata
-
- .align ALIGNARG(4)
- ASM_TYPE_DIRECTIVE(huge_nan_null_null,@object)
-huge_nan_null_null:
- .byte 0, 0, 0, 0, 0, 0, 0xf0, 0x7f
- .byte 0, 0, 0, 0, 0, 0, 0xff, 0x7f
- .double 0.0
-zero: .double 0.0
-infinity:
- .byte 0, 0, 0, 0, 0, 0, 0xf0, 0x7f
- .byte 0, 0, 0, 0, 0, 0, 0xff, 0x7f
- .double 0.0
- .byte 0, 0, 0, 0, 0, 0, 0, 0x80
- ASM_SIZE_DIRECTIVE(huge_nan_null_null)
-
- ASM_TYPE_DIRECTIVE(twopi,@object)
-twopi:
- .byte 0x35, 0xc2, 0x68, 0x21, 0xa2, 0xda, 0xf, 0xc9, 0x1, 0x40
- .byte 0, 0, 0, 0, 0, 0
- ASM_SIZE_DIRECTIVE(twopi)
-
- ASM_TYPE_DIRECTIVE(l2e,@object)
-l2e:
- .byte 0xbc, 0xf0, 0x17, 0x5c, 0x29, 0x3b, 0xaa, 0xb8, 0xff, 0x3f
- .byte 0, 0, 0, 0, 0, 0
- ASM_SIZE_DIRECTIVE(l2e)
-
- ASM_TYPE_DIRECTIVE(one,@object)
-one: .double 1.0
- ASM_SIZE_DIRECTIVE(one)
-
-
-#ifdef PIC
-#define MO(op) op##@GOTOFF(%ecx)
-#define MOX(op,x,f) op##@GOTOFF(%ecx,x,f)
-#else
-#define MO(op) op
-#define MOX(op,x,f) op(,x,f)
-#endif
-
- .text
-ENTRY(__cexpl)
- fldt 8(%esp) /* x */
- fxam
- fnstsw
- fldt 20(%esp) /* y : x */
-#ifdef PIC
- LOAD_PIC_REG (cx)
-#endif
- movb %ah, %dh
- andb $0x45, %ah
- cmpb $0x05, %ah
- je 1f /* Jump if real part is +-Inf */
- cmpb $0x01, %ah
- je 2f /* Jump if real part is NaN */
-
- fxam /* y : x */
- fnstsw
- /* If the imaginary part is not finite we return NaN+i NaN, as
- for the case when the real part is NaN. A test for +-Inf and
- NaN would be necessary. But since we know the stack register
- we applied `fxam' to is not empty we can simply use one test.
- Check your FPU manual for more information. */
- andb $0x01, %ah
- cmpb $0x01, %ah
- je 20f
-
- /* We have finite numbers in the real and imaginary part. Do
- the real work now. */
- fxch /* x : y */
- fldt MO(l2e) /* log2(e) : x : y */
- fmulp /* x * log2(e) : y */
- fld %st /* x * log2(e) : x * log2(e) : y */
- frndint /* int(x * log2(e)) : x * log2(e) : y */
- fsubr %st, %st(1) /* int(x * log2(e)) : frac(x * log2(e)) : y */
- fxch /* frac(x * log2(e)) : int(x * log2(e)) : y */
- f2xm1 /* 2^frac(x * log2(e))-1 : int(x * log2(e)) : y */
- faddl MO(one) /* 2^frac(x * log2(e)) : int(x * log2(e)) : y */
- fscale /* e^x : int(x * log2(e)) : y */
- fst %st(1) /* e^x : e^x : y */
- fxch %st(2) /* y : e^x : e^x */
- fsincos /* cos(y) : sin(y) : e^x : e^x */
- fnstsw
- testl $0x400, %eax
- jnz 7f
- fmulp %st, %st(3) /* sin(y) : e^x : e^x * cos(y) */
- fmulp %st, %st(1) /* e^x * sin(y) : e^x * cos(y) */
- movl 4(%esp), %eax /* Pointer to memory for result. */
- fstpt 12(%eax)
- fstpt (%eax)
- ret $4
-
- /* We have to reduce the argument to fsincos. */
- .align ALIGNARG(4)
-7: fldt MO(twopi) /* 2*pi : y : e^x : e^x */
- fxch /* y : 2*pi : e^x : e^x */
-8: fprem1 /* y%(2*pi) : 2*pi : e^x : e^x */
- fnstsw
- testl $0x400, %eax
- jnz 8b
- fstp %st(1) /* y%(2*pi) : e^x : e^x */
- fsincos /* cos(y) : sin(y) : e^x : e^x */
- fmulp %st, %st(3)
- fmulp %st, %st(1)
- movl 4(%esp), %eax /* Pointer to memory for result. */
- fstpt 12(%eax)
- fstpt (%eax)
- ret $4
-
- /* The real part is +-inf. We must make further differences. */
- .align ALIGNARG(4)
-1: fxam /* y : x */
- fnstsw
- movb %ah, %dl
- testb $0x01, %ah /* See above why 0x01 is usable here. */
- jne 3f
-
-
- /* The real part is +-Inf and the imaginary part is finite. */
- andl $0x245, %edx
- cmpb $0x40, %dl /* Imaginary part == 0? */
- je 4f /* Yes -> */
-
- fxch /* x : y */
- shrl $5, %edx
- fstp %st(0) /* y */ /* Drop the real part. */
- andl $16, %edx /* This puts the sign bit of the real part
- in bit 4. So we can use it to index a
- small array to select 0 or Inf. */
- fsincos /* cos(y) : sin(y) */
- fnstsw
- testl $0x0400, %eax
- jnz 5f
- fldl MOX(huge_nan_null_null,%edx,1)
- movl 4(%esp), %edx /* Pointer to memory for result. */
- fld %st
- fstpt 12(%edx)
- fstpt (%edx)
- ftst
- fnstsw
- shll $7, %eax
- andl $0x8000, %eax
- orl %eax, 8(%edx)
- fstp %st(0)
- ftst
- fnstsw
- shll $7, %eax
- andl $0x8000, %eax
- orl %eax, 20(%edx)
- fstp %st(0)
- ret $4
- /* We must reduce the argument to fsincos. */
- .align ALIGNARG(4)
-5: fldt MO(twopi)
- fxch
-6: fprem1
- fnstsw
- testl $0x400, %eax
- jnz 6b
- fstp %st(1)
- fsincos
- fldl MOX(huge_nan_null_null,%edx,1)
- movl 4(%esp), %edx /* Pointer to memory for result. */
- fld %st
- fstpt 12(%edx)
- fstpt (%edx)
- ftst
- fnstsw
- shll $7, %eax
- andl $0x8000, %eax
- orl %eax, 8(%edx)
- fstp %st(0)
- ftst
- fnstsw
- shll $7, %eax
- andl $0x8000, %eax
- orl %eax, 20(%edx)
- fstp %st(0)
- ret $4
-
- /* The real part is +-Inf and the imaginary part is +-0. So return
- +-Inf+-0i. */
- .align ALIGNARG(4)
-4: movl 4(%esp), %eax /* Pointer to memory for result. */
- fstpt 12(%eax)
- shrl $5, %edx
- fstp %st(0)
- andl $16, %edx
- fldl MOX(huge_nan_null_null,%edx,1)
- fstpt (%eax)
- ret $4
-
- /* The real part is +-Inf, the imaginary is also is not finite. */
- .align ALIGNARG(4)
-3: fstp %st(0)
- fstp %st(0) /* <empty> */
- andb $0x45, %ah
- andb $0x47, %dh
- xorb %dh, %ah
- jnz 30f
- fldl MO(infinity) /* Raise invalid exception. */
- fmull MO(zero)
- fstp %st(0)
-30: movl %edx, %eax
- shrl $5, %edx
- shll $4, %eax
- andl $16, %edx
- andl $32, %eax
- orl %eax, %edx
- movl 4(%esp), %eax /* Pointer to memory for result. */
-
- fldl MOX(huge_nan_null_null,%edx,1)
- fldl MOX(huge_nan_null_null+8,%edx,1)
- fxch
- fstpt (%eax)
- fstpt 12(%eax)
- ret $4
-
- /* The real part is NaN. */
- .align ALIGNARG(4)
-20: fldl MO(infinity) /* Raise invalid exception. */
- fmull MO(zero)
- fstp %st(0)
-2: fstp %st(0)
- fstp %st(0)
- movl 4(%esp), %eax /* Pointer to memory for result. */
- fldl MO(huge_nan_null_null+8)
- fld %st(0)
- fstpt (%eax)
- fstpt 12(%eax)
- ret $4
-
-END(__cexpl)
-weak_alias (__cexpl, cexpl)
Modified: fsf/trunk/libc/sysdeps/ieee754/dbl-64/e_pow.c
==============================================================================
--- fsf/trunk/libc/sysdeps/ieee754/dbl-64/e_pow.c (original)
+++ fsf/trunk/libc/sysdeps/ieee754/dbl-64/e_pow.c Wed Mar 21 19:50:58 2012
@@ -111,7 +111,7 @@
if (((v.i[HIGH_HALF] & 0x7fffffff) == 0x7ff00000 && v.i[LOW_HALF] != 0)
|| (v.i[HIGH_HALF] & 0x7fffffff) > 0x7ff00000)
return y;
- if (ABS(y) > 1.0e20) return (y>0)?0:INF.x;
+ if (ABS(y) > 1.0e20) return (y>0)?0:1.0/ABS(x);
k = checkint(y);
if (k == -1)
return y < 0 ? 1.0/x : x;
Modified: fsf/trunk/libc/sysdeps/unix/sysv/linux/configure
==============================================================================
--- fsf/trunk/libc/sysdeps/unix/sysv/linux/configure (original)
+++ fsf/trunk/libc/sysdeps/unix/sysv/linux/configure Wed Mar 21 19:50:58 2012
@@ -393,7 +393,7 @@
# and libc_cv_localedir.
test -n "$libc_cv_slibdir" || \
case $machine in
- sparc/sparc64 | x86_64 | powerpc/powerpc64 | s390/s390-64)
+ sparc/sparc64 | x86_64* | powerpc/powerpc64 | s390/s390-64)
libc_cv_slibdir="/lib64"
if test "$libdir" = '${exec_prefix}/lib'; then
libdir='${exec_prefix}/lib64';
Modified: fsf/trunk/libc/sysdeps/unix/sysv/linux/configure.in
==============================================================================
--- fsf/trunk/libc/sysdeps/unix/sysv/linux/configure.in (original)
+++ fsf/trunk/libc/sysdeps/unix/sysv/linux/configure.in Wed Mar 21 19:50:58 2012
@@ -132,7 +132,7 @@
# and libc_cv_localedir.
test -n "$libc_cv_slibdir" || \
case $machine in
- sparc/sparc64 | x86_64 | powerpc/powerpc64 | s390/s390-64)
+ sparc/sparc64 | x86_64* | powerpc/powerpc64 | s390/s390-64)
libc_cv_slibdir="/lib64"
if test "$libdir" = '${exec_prefix}/lib'; then
libdir='${exec_prefix}/lib64';
Modified: fsf/trunk/libc/sysdeps/x86_64/fpu/e_powl.S
==============================================================================
--- fsf/trunk/libc/sysdeps/x86_64/fpu/e_powl.S (original)
+++ fsf/trunk/libc/sysdeps/x86_64/fpu/e_powl.S Wed Mar 21 19:50:58 2012
@@ -32,6 +32,9 @@
ASM_TYPE_DIRECTIVE(p63,@object)
p63: .byte 0, 0, 0, 0, 0, 0, 0xe0, 0x43
ASM_SIZE_DIRECTIVE(p63)
+ ASM_TYPE_DIRECTIVE(p64,@object)
+p64: .byte 0, 0, 0, 0, 0, 0, 0xf0, 0x43
+ ASM_SIZE_DIRECTIVE(p64)
.section .rodata.cst16,"aM",@progbits,16
@@ -227,6 +230,19 @@
testb $2, %dh
jz 16f // jump if x == +inf
+ // fistpll raises invalid exception for |y| >= 1L<<63, but y
+ // may be odd unless we know |y| >= 1L<<64.
+ fldl MO(p64) // 1L<<64 : y
+ fld %st(1) // y : 1L<<64 : y
+ fabs // |y| : 1L<<64 : y
+ fcomip %st(1), %st // 1L<<64 : y
+ fstp %st(0) // y
+ jnc 16f
+ fldl MO(p63) // p63 : y
+ fxch // y : p63
+ fprem // y%p63 : p63
+ fstp %st(1) // y%p63
+
// We must find out whether y is an odd integer.
fld %st // y : y
fistpll -8(%rsp) // y
@@ -283,6 +299,19 @@
// x is Â0 and y is < 0. We must find out whether y is an odd integer.
testb $2, %dh
jz 25f
+
+ // fistpll raises invalid exception for |y| >= 1L<<63, but y
+ // may be odd unless we know |y| >= 1L<<64.
+ fldl MO(p64) // 1L<<64 : y
+ fld %st(1) // y : 1L<<64 : y
+ fabs // |y| : 1L<<64 : y
+ fcomip %st(1), %st // 1L<<64 : y
+ fstp %st(0) // y
+ jnc 25f
+ fldl MO(p63) // p63 : y
+ fxch // y : p63
+ fprem // y%p63 : p63
+ fstp %st(1) // y%p63
fld %st // y : y
fistpll -8(%rsp) // y
@@ -315,6 +344,18 @@
21: testb $2, %dh
jz 22f
+ // fistpll raises invalid exception for |y| >= 1L<<63, but y
+ // may be odd unless we know |y| >= 1L<<64.
+ fldl MO(p64) // 1L<<64 : y
+ fxch // y : 1L<<64
+ fcomi %st(1), %st // y : 1L<<64
+ fstp %st(1) // y
+ jnc 22f
+ fldl MO(p63) // p63 : y
+ fxch // y : p63
+ fprem // y%p63 : p63
+ fstp %st(1) // y%p63
+
fld %st // y : y
fistpll -8(%rsp) // y
fildll -8(%rsp) // int(y) : y
Modified: fsf/trunk/libc/sysdeps/x86_64/fpu/libm-test-ulps
==============================================================================
--- fsf/trunk/libc/sysdeps/x86_64/fpu/libm-test-ulps (original)
+++ fsf/trunk/libc/sysdeps/x86_64/fpu/libm-test-ulps Wed Mar 21 19:50:58 2012
@@ -482,6 +482,9 @@
ifloat: 1
# cexp
+Test "Real part of: cexp (-10000 + 0x1p16383 i) == 1.045876464564882298442774542991176546722e-4343 + 4.421154026488516836023811173959413420548e-4344 i":
+ildouble: 1
+ldouble: 1
Test "Imaginary part of: cexp (-2.0 - 3.0 i) == -0.13398091492954261346140525546115575 - 0.019098516261135196432576240858800925 i":
float: 1
ifloat: 1
@@ -491,6 +494,19 @@
Test "Imaginary part of: cexp (0.75 + 1.25 i) == 0.667537446429131586942201977015932112 + 2.00900045494094876258347228145863909 i":
ildouble: 1
ldouble: 1
+Test "Real part of: cexp (50 + 0x1p127 i) == 4.053997150228616856622417636046265337193e21 + 3.232070315463388524466674772633810238819e21 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+Test "Imaginary part of: cexp (50 + 0x1p127 i) == 4.053997150228616856622417636046265337193e21 + 3.232070315463388524466674772633810238819e21 i":
+double: 1
+idouble: 1
+ildouble: 1
+ldouble: 1
+Test "Real part of: cexp (500 + 0x1p1023 i) == -1.159886268932754433233243794561351783426e217 + 7.904017694554466595359379965081774849708e216 i":
+double: 1
+idouble: 1
# clog
Test "Imaginary part of: clog (-2 - 3 i) == 1.2824746787307683680267437207826593 - 2.1587989303424641704769327722648368 i":
@@ -2090,11 +2106,17 @@
ldouble: 1
Function: Real part of "cexp":
-float: 1
-ifloat: 1
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+ildouble: 1
+ldouble: 1
Function: Imaginary part of "cexp":
-float: 1
+double: 1
+float: 1
+idouble: 1
ifloat: 1
ildouble: 1
ldouble: 1
Modified: fsf/trunk/libc/timezone/Makefile
==============================================================================
--- fsf/trunk/libc/timezone/Makefile (original)
+++ fsf/trunk/libc/timezone/Makefile Wed Mar 21 19:50:58 2012
@@ -45,7 +45,6 @@
CPPFLAGS-zic = -DNOT_IN_libc
ifeq ($(have-ksh),yes)
-install-others += $(inst_zonedir)/iso3166.tab $(inst_zonedir)/zone.tab
install-bin-script = tzselect
generated += tzselect
endif
@@ -111,7 +110,3 @@
-e 's%@TZDIR@%$(zonedir)%g' < $< > $@.new
chmod 555 $@.new
mv -f $@.new $@
-
-$(addprefix $(inst_zonedir)/,iso3166.tab zone.tab): \
- $(inst_zonedir)/%: % $(+force)
- $(do-install)
_______________________________________________
Commits mailing list
Commits@xxxxxxxxxx
http://eglibc.org/cgi-bin/mailman/listinfo/commits